W3.8: Investigation Bureau complete — Poirot, Taleb, Tetlock, Case-Writer
Brings the bureau from 4 → 8 detectives. All eight run as Bun + claude-CLI
subprocesses against the same Supabase + investigation_jobs LISTEN/NOTIFY
queue, sharing search.ts hybridSearch and writer-side validators that
gate writes against schema + FK.
New detectives:
Poirot (witness_analysis)
- prompts/poirot.md — credibility / access / bias / corroboration /
verdict; uses entity_mentions JOIN chunks to pull 12 chunks per
person; resolves corroboration_refs chunk_ids defensively (accepts
bare cNNNN even when the model emits pNNN/cNNNN).
- INSERT into public.witnesses with W-NNNN naming.
- Tone: purple (#9b5de5).
Taleb (outlier_scan)
- prompts/taleb.md — "surprise is relative to a model"; at most 3
outliers; each requires explicit dominant_model + why_surprising +
what_it_implies; fan-out into public.gaps with scope.kind="outlier".
- Same unscoped-fallback as Dupin (Pass 1 with doc_id, Pass 2 widens
to corpus if hits < 3).
- Tone: yellow (#ffd23f).
Tetlock (calibrate_hypothesis)
- prompts/tetlock.md — honest Bayesian update; emits new_posterior +
Δ + recommended_action ∈ {keep, downgrade, upgrade, supersede}.
- write_calibration UPDATEs public.hypotheses + APPENDS a
"## Calibration history" section to the H-NNNN.md case file
(calibration is append-only — each datapoint matters). Posterior
band auto-corrected to match Tetlock thresholds.
- NO_NEW_EVIDENCE sentinel handled; pure 'keep' with |Δ|<0.005 only
touches updated_at + reviewed_by.
- Tone: teal (#26d4cc).
Case-Writer (case_report)
- prompts/case-writer.md — Dr. Watson assembles all artefacts
(E-NNNN, H-NNNN, R-NNNN, W-NNNN, G-NNNN) into a five-act narrative.
ILIKE filter on topic; doc_id optional scope.
- Larger budget cap (≥ $0.50) + longer timeout for prose generation.
- Writes case/reports/<slug>.md with frontmatter (topic + counts);
no DB table for v0.
- New page /c/[slug] renders the report via MarkdownBody + stat chips.
- Tone: gold (#e0c080).
Hardening across the bureau:
- Sentinel parsing now accepts backticked AND prose-trailing forms
(Holmes NO_HYPOTHESES, Dupin NO_CONTRADICTIONS, Schneier
INSUFFICIENT_HYPOTHESIS, Poirot INSUFFICIENT_TESTIMONY, Taleb
NO_OUTLIERS, Tetlock NO_NEW_EVIDENCE, Case-Writer
INSUFFICIENT_ARTEFACTS). Avoids the failure mode where the model
refuses honestly but the runtime treated it as a parse error
(observed live with Poirot+Hoover identifying the DIRECTOR
false-positive disambiguation issue in entity_mentions).
Chat tool extensions (web/lib/chat/tools.ts):
- request_investigation now accepts 7 kinds. Each routes to its
detective with appropriate validation (hypothesis_id regex,
person_id kebab-case, topic non-empty, doc_id for evidence_chain).
- ETA per kind: Holmes/Dupin 60s, Poirot 45s, Schneier/Tetlock 30s,
Taleb 50s, Case-Writer 180s (longer prose), Locard 30×n_chunks.
UI integration:
- chat-bubble inline card paints each detective in its tone color.
- /jobs/[id] page header swaps name/subtitle/tone per detective;
question label adapts ("Topic" / "Hypothesis under attack" /
"Witness under analysis" / "Topic to outlier-scan" / "Hypothesis
under recalibration" / "Case to assemble").
- job-status-poller renders: case-report link card (gold), outlier
cards (yellow), witness cards (purple) — alongside existing
hypothesis, evidence, contradiction cards.
- /api/jobs/[id] hydrates witnesses (JOIN entities for canonical_name)
+ gaps (with scope JSONB).
- /c/[slug] page reads /data/ufo/case/reports/<slug>.md and renders
with MarkdownBody, frontmatter parsed for stat chips.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 01:11:39 +00:00
|
|
|
/**
|
W5.4 (Phase 3B): sitemap + robots + Article schema + magazine reading view
GEO/SEO surface area:
app/robots.ts (new — Next.js dynamic robots)
Explicitly ALLOWS major AI crawlers: GPTBot, OAI-SearchBot,
ChatGPT-User, ClaudeBot, Claude-Web, anthropic-ai, PerplexityBot,
Perplexity-User, Google-Extended, Applebot-Extended, CCBot,
DuckAssistBot, YouBot, Bytespider, Amazonbot. The site exists to
be cited by LLMs answering UAP/UFO questions — we want them in.
/api/admin/, /admin/, /auth/ disallowed for everyone.
app/sitemap.ts (new — Next.js dynamic sitemap)
Lists 9 top-level routes + every /d/<doc> + every /c/<slug> from
the filesystem + up to 500 entity URLs per class
(event, person, uap_object, location, organization),
sorted with summary-enriched entities first. ~3000 URLs total at
current corpus size. lastModified honours summary_generated_at so
crawlers re-index when entities are re-enriched.
app/c/[slug]/page.tsx (rewritten — magazine reading view)
- generateMetadata: per-case title, description (auto-extracted
from the locale-preferred lead paragraph), canonical URL,
hreflang alternate, OpenGraph article type with publishedTime,
Twitter card.
- JSON-LD Article schema embedded at end of page: schema.org
Article + Organization publisher + inLanguage + isAccessibleForFree.
This is what makes the case appear as a citable source in
Google AI Overviews / Perplexity / ChatGPT search.
- Reading view rewritten: display-serif headline (Fraunces), italic
blockquotes with gold accent, prose-typography styling, no more
detective stats line, no more "written by case-writer@detective"
attribution. Locale-aware: PT-BR pulls topic_pt_br + lead in PT,
English mirror.
tailwind.config.ts
+ @tailwindcss/typography plugin
+ font-display family wired to var(--font-display) (Fraunces)
package.json
+ "@tailwindcss/typography" devDependency
Phase 3A note: bulk entity enrichment hit Claude OAuth weekly quota mid-run.
6 events + 3 uap_objects landed bilingual summaries before the quota
exhausted. UI gracefully splits enriched vs bare entities so /sightings
shows the magazine-grade cards (Kenneth Arnold 1947, Roswell, Maury Island,
Joseph Perry 1960 lunar photo, Civil Defense Director 1966, etc.) on top
of a compact table of the rest. Re-run when quota refreshes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 19:09:50 +00:00
|
|
|
* /c/[slug] — Case file reader.
|
W3.8: Investigation Bureau complete — Poirot, Taleb, Tetlock, Case-Writer
Brings the bureau from 4 → 8 detectives. All eight run as Bun + claude-CLI
subprocesses against the same Supabase + investigation_jobs LISTEN/NOTIFY
queue, sharing search.ts hybridSearch and writer-side validators that
gate writes against schema + FK.
New detectives:
Poirot (witness_analysis)
- prompts/poirot.md — credibility / access / bias / corroboration /
verdict; uses entity_mentions JOIN chunks to pull 12 chunks per
person; resolves corroboration_refs chunk_ids defensively (accepts
bare cNNNN even when the model emits pNNN/cNNNN).
- INSERT into public.witnesses with W-NNNN naming.
- Tone: purple (#9b5de5).
Taleb (outlier_scan)
- prompts/taleb.md — "surprise is relative to a model"; at most 3
outliers; each requires explicit dominant_model + why_surprising +
what_it_implies; fan-out into public.gaps with scope.kind="outlier".
- Same unscoped-fallback as Dupin (Pass 1 with doc_id, Pass 2 widens
to corpus if hits < 3).
- Tone: yellow (#ffd23f).
Tetlock (calibrate_hypothesis)
- prompts/tetlock.md — honest Bayesian update; emits new_posterior +
Δ + recommended_action ∈ {keep, downgrade, upgrade, supersede}.
- write_calibration UPDATEs public.hypotheses + APPENDS a
"## Calibration history" section to the H-NNNN.md case file
(calibration is append-only — each datapoint matters). Posterior
band auto-corrected to match Tetlock thresholds.
- NO_NEW_EVIDENCE sentinel handled; pure 'keep' with |Δ|<0.005 only
touches updated_at + reviewed_by.
- Tone: teal (#26d4cc).
Case-Writer (case_report)
- prompts/case-writer.md — Dr. Watson assembles all artefacts
(E-NNNN, H-NNNN, R-NNNN, W-NNNN, G-NNNN) into a five-act narrative.
ILIKE filter on topic; doc_id optional scope.
- Larger budget cap (≥ $0.50) + longer timeout for prose generation.
- Writes case/reports/<slug>.md with frontmatter (topic + counts);
no DB table for v0.
- New page /c/[slug] renders the report via MarkdownBody + stat chips.
- Tone: gold (#e0c080).
Hardening across the bureau:
- Sentinel parsing now accepts backticked AND prose-trailing forms
(Holmes NO_HYPOTHESES, Dupin NO_CONTRADICTIONS, Schneier
INSUFFICIENT_HYPOTHESIS, Poirot INSUFFICIENT_TESTIMONY, Taleb
NO_OUTLIERS, Tetlock NO_NEW_EVIDENCE, Case-Writer
INSUFFICIENT_ARTEFACTS). Avoids the failure mode where the model
refuses honestly but the runtime treated it as a parse error
(observed live with Poirot+Hoover identifying the DIRECTOR
false-positive disambiguation issue in entity_mentions).
Chat tool extensions (web/lib/chat/tools.ts):
- request_investigation now accepts 7 kinds. Each routes to its
detective with appropriate validation (hypothesis_id regex,
person_id kebab-case, topic non-empty, doc_id for evidence_chain).
- ETA per kind: Holmes/Dupin 60s, Poirot 45s, Schneier/Tetlock 30s,
Taleb 50s, Case-Writer 180s (longer prose), Locard 30×n_chunks.
UI integration:
- chat-bubble inline card paints each detective in its tone color.
- /jobs/[id] page header swaps name/subtitle/tone per detective;
question label adapts ("Topic" / "Hypothesis under attack" /
"Witness under analysis" / "Topic to outlier-scan" / "Hypothesis
under recalibration" / "Case to assemble").
- job-status-poller renders: case-report link card (gold), outlier
cards (yellow), witness cards (purple) — alongside existing
hypothesis, evidence, contradiction cards.
- /api/jobs/[id] hydrates witnesses (JOIN entities for canonical_name)
+ gaps (with scope JSONB).
- /c/[slug] page reads /data/ufo/case/reports/<slug>.md and renders
with MarkdownBody, frontmatter parsed for stat chips.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 01:11:39 +00:00
|
|
|
*
|
W5.4 (Phase 3B): sitemap + robots + Article schema + magazine reading view
GEO/SEO surface area:
app/robots.ts (new — Next.js dynamic robots)
Explicitly ALLOWS major AI crawlers: GPTBot, OAI-SearchBot,
ChatGPT-User, ClaudeBot, Claude-Web, anthropic-ai, PerplexityBot,
Perplexity-User, Google-Extended, Applebot-Extended, CCBot,
DuckAssistBot, YouBot, Bytespider, Amazonbot. The site exists to
be cited by LLMs answering UAP/UFO questions — we want them in.
/api/admin/, /admin/, /auth/ disallowed for everyone.
app/sitemap.ts (new — Next.js dynamic sitemap)
Lists 9 top-level routes + every /d/<doc> + every /c/<slug> from
the filesystem + up to 500 entity URLs per class
(event, person, uap_object, location, organization),
sorted with summary-enriched entities first. ~3000 URLs total at
current corpus size. lastModified honours summary_generated_at so
crawlers re-index when entities are re-enriched.
app/c/[slug]/page.tsx (rewritten — magazine reading view)
- generateMetadata: per-case title, description (auto-extracted
from the locale-preferred lead paragraph), canonical URL,
hreflang alternate, OpenGraph article type with publishedTime,
Twitter card.
- JSON-LD Article schema embedded at end of page: schema.org
Article + Organization publisher + inLanguage + isAccessibleForFree.
This is what makes the case appear as a citable source in
Google AI Overviews / Perplexity / ChatGPT search.
- Reading view rewritten: display-serif headline (Fraunces), italic
blockquotes with gold accent, prose-typography styling, no more
detective stats line, no more "written by case-writer@detective"
attribution. Locale-aware: PT-BR pulls topic_pt_br + lead in PT,
English mirror.
tailwind.config.ts
+ @tailwindcss/typography plugin
+ font-display family wired to var(--font-display) (Fraunces)
package.json
+ "@tailwindcss/typography" devDependency
Phase 3A note: bulk entity enrichment hit Claude OAuth weekly quota mid-run.
6 events + 3 uap_objects landed bilingual summaries before the quota
exhausted. UI gracefully splits enriched vs bare entities so /sightings
shows the magazine-grade cards (Kenneth Arnold 1947, Roswell, Maury Island,
Joseph Perry 1960 lunar photo, Civil Defense Director 1966, etc.) on top
of a compact table of the rest. Re-run when quota refreshes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 19:09:50 +00:00
|
|
|
* Renders a single narrated case file from /data/ufo/case/reports/<slug>.md.
|
|
|
|
|
* The reader sees a magazine-style article — title, dateline, body. No
|
|
|
|
|
* detective attribution, no skeptic framing.
|
W3.8: Investigation Bureau complete — Poirot, Taleb, Tetlock, Case-Writer
Brings the bureau from 4 → 8 detectives. All eight run as Bun + claude-CLI
subprocesses against the same Supabase + investigation_jobs LISTEN/NOTIFY
queue, sharing search.ts hybridSearch and writer-side validators that
gate writes against schema + FK.
New detectives:
Poirot (witness_analysis)
- prompts/poirot.md — credibility / access / bias / corroboration /
verdict; uses entity_mentions JOIN chunks to pull 12 chunks per
person; resolves corroboration_refs chunk_ids defensively (accepts
bare cNNNN even when the model emits pNNN/cNNNN).
- INSERT into public.witnesses with W-NNNN naming.
- Tone: purple (#9b5de5).
Taleb (outlier_scan)
- prompts/taleb.md — "surprise is relative to a model"; at most 3
outliers; each requires explicit dominant_model + why_surprising +
what_it_implies; fan-out into public.gaps with scope.kind="outlier".
- Same unscoped-fallback as Dupin (Pass 1 with doc_id, Pass 2 widens
to corpus if hits < 3).
- Tone: yellow (#ffd23f).
Tetlock (calibrate_hypothesis)
- prompts/tetlock.md — honest Bayesian update; emits new_posterior +
Δ + recommended_action ∈ {keep, downgrade, upgrade, supersede}.
- write_calibration UPDATEs public.hypotheses + APPENDS a
"## Calibration history" section to the H-NNNN.md case file
(calibration is append-only — each datapoint matters). Posterior
band auto-corrected to match Tetlock thresholds.
- NO_NEW_EVIDENCE sentinel handled; pure 'keep' with |Δ|<0.005 only
touches updated_at + reviewed_by.
- Tone: teal (#26d4cc).
Case-Writer (case_report)
- prompts/case-writer.md — Dr. Watson assembles all artefacts
(E-NNNN, H-NNNN, R-NNNN, W-NNNN, G-NNNN) into a five-act narrative.
ILIKE filter on topic; doc_id optional scope.
- Larger budget cap (≥ $0.50) + longer timeout for prose generation.
- Writes case/reports/<slug>.md with frontmatter (topic + counts);
no DB table for v0.
- New page /c/[slug] renders the report via MarkdownBody + stat chips.
- Tone: gold (#e0c080).
Hardening across the bureau:
- Sentinel parsing now accepts backticked AND prose-trailing forms
(Holmes NO_HYPOTHESES, Dupin NO_CONTRADICTIONS, Schneier
INSUFFICIENT_HYPOTHESIS, Poirot INSUFFICIENT_TESTIMONY, Taleb
NO_OUTLIERS, Tetlock NO_NEW_EVIDENCE, Case-Writer
INSUFFICIENT_ARTEFACTS). Avoids the failure mode where the model
refuses honestly but the runtime treated it as a parse error
(observed live with Poirot+Hoover identifying the DIRECTOR
false-positive disambiguation issue in entity_mentions).
Chat tool extensions (web/lib/chat/tools.ts):
- request_investigation now accepts 7 kinds. Each routes to its
detective with appropriate validation (hypothesis_id regex,
person_id kebab-case, topic non-empty, doc_id for evidence_chain).
- ETA per kind: Holmes/Dupin 60s, Poirot 45s, Schneier/Tetlock 30s,
Taleb 50s, Case-Writer 180s (longer prose), Locard 30×n_chunks.
UI integration:
- chat-bubble inline card paints each detective in its tone color.
- /jobs/[id] page header swaps name/subtitle/tone per detective;
question label adapts ("Topic" / "Hypothesis under attack" /
"Witness under analysis" / "Topic to outlier-scan" / "Hypothesis
under recalibration" / "Case to assemble").
- job-status-poller renders: case-report link card (gold), outlier
cards (yellow), witness cards (purple) — alongside existing
hypothesis, evidence, contradiction cards.
- /api/jobs/[id] hydrates witnesses (JOIN entities for canonical_name)
+ gaps (with scope JSONB).
- /c/[slug] page reads /data/ufo/case/reports/<slug>.md and renders
with MarkdownBody, frontmatter parsed for stat chips.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 01:11:39 +00:00
|
|
|
*/
|
|
|
|
|
import { notFound } from "next/navigation";
|
|
|
|
|
import { readFile } from "node:fs/promises";
|
|
|
|
|
import path from "node:path";
|
W5.4 (Phase 3B): sitemap + robots + Article schema + magazine reading view
GEO/SEO surface area:
app/robots.ts (new — Next.js dynamic robots)
Explicitly ALLOWS major AI crawlers: GPTBot, OAI-SearchBot,
ChatGPT-User, ClaudeBot, Claude-Web, anthropic-ai, PerplexityBot,
Perplexity-User, Google-Extended, Applebot-Extended, CCBot,
DuckAssistBot, YouBot, Bytespider, Amazonbot. The site exists to
be cited by LLMs answering UAP/UFO questions — we want them in.
/api/admin/, /admin/, /auth/ disallowed for everyone.
app/sitemap.ts (new — Next.js dynamic sitemap)
Lists 9 top-level routes + every /d/<doc> + every /c/<slug> from
the filesystem + up to 500 entity URLs per class
(event, person, uap_object, location, organization),
sorted with summary-enriched entities first. ~3000 URLs total at
current corpus size. lastModified honours summary_generated_at so
crawlers re-index when entities are re-enriched.
app/c/[slug]/page.tsx (rewritten — magazine reading view)
- generateMetadata: per-case title, description (auto-extracted
from the locale-preferred lead paragraph), canonical URL,
hreflang alternate, OpenGraph article type with publishedTime,
Twitter card.
- JSON-LD Article schema embedded at end of page: schema.org
Article + Organization publisher + inLanguage + isAccessibleForFree.
This is what makes the case appear as a citable source in
Google AI Overviews / Perplexity / ChatGPT search.
- Reading view rewritten: display-serif headline (Fraunces), italic
blockquotes with gold accent, prose-typography styling, no more
detective stats line, no more "written by case-writer@detective"
attribution. Locale-aware: PT-BR pulls topic_pt_br + lead in PT,
English mirror.
tailwind.config.ts
+ @tailwindcss/typography plugin
+ font-display family wired to var(--font-display) (Fraunces)
package.json
+ "@tailwindcss/typography" devDependency
Phase 3A note: bulk entity enrichment hit Claude OAuth weekly quota mid-run.
6 events + 3 uap_objects landed bilingual summaries before the quota
exhausted. UI gracefully splits enriched vs bare entities so /sightings
shows the magazine-grade cards (Kenneth Arnold 1947, Roswell, Maury Island,
Joseph Perry 1960 lunar photo, Civil Defense Director 1966, etc.) on top
of a compact table of the rest. Re-run when quota refreshes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 19:09:50 +00:00
|
|
|
import type { Metadata } from "next";
|
W3.8: Investigation Bureau complete — Poirot, Taleb, Tetlock, Case-Writer
Brings the bureau from 4 → 8 detectives. All eight run as Bun + claude-CLI
subprocesses against the same Supabase + investigation_jobs LISTEN/NOTIFY
queue, sharing search.ts hybridSearch and writer-side validators that
gate writes against schema + FK.
New detectives:
Poirot (witness_analysis)
- prompts/poirot.md — credibility / access / bias / corroboration /
verdict; uses entity_mentions JOIN chunks to pull 12 chunks per
person; resolves corroboration_refs chunk_ids defensively (accepts
bare cNNNN even when the model emits pNNN/cNNNN).
- INSERT into public.witnesses with W-NNNN naming.
- Tone: purple (#9b5de5).
Taleb (outlier_scan)
- prompts/taleb.md — "surprise is relative to a model"; at most 3
outliers; each requires explicit dominant_model + why_surprising +
what_it_implies; fan-out into public.gaps with scope.kind="outlier".
- Same unscoped-fallback as Dupin (Pass 1 with doc_id, Pass 2 widens
to corpus if hits < 3).
- Tone: yellow (#ffd23f).
Tetlock (calibrate_hypothesis)
- prompts/tetlock.md — honest Bayesian update; emits new_posterior +
Δ + recommended_action ∈ {keep, downgrade, upgrade, supersede}.
- write_calibration UPDATEs public.hypotheses + APPENDS a
"## Calibration history" section to the H-NNNN.md case file
(calibration is append-only — each datapoint matters). Posterior
band auto-corrected to match Tetlock thresholds.
- NO_NEW_EVIDENCE sentinel handled; pure 'keep' with |Δ|<0.005 only
touches updated_at + reviewed_by.
- Tone: teal (#26d4cc).
Case-Writer (case_report)
- prompts/case-writer.md — Dr. Watson assembles all artefacts
(E-NNNN, H-NNNN, R-NNNN, W-NNNN, G-NNNN) into a five-act narrative.
ILIKE filter on topic; doc_id optional scope.
- Larger budget cap (≥ $0.50) + longer timeout for prose generation.
- Writes case/reports/<slug>.md with frontmatter (topic + counts);
no DB table for v0.
- New page /c/[slug] renders the report via MarkdownBody + stat chips.
- Tone: gold (#e0c080).
Hardening across the bureau:
- Sentinel parsing now accepts backticked AND prose-trailing forms
(Holmes NO_HYPOTHESES, Dupin NO_CONTRADICTIONS, Schneier
INSUFFICIENT_HYPOTHESIS, Poirot INSUFFICIENT_TESTIMONY, Taleb
NO_OUTLIERS, Tetlock NO_NEW_EVIDENCE, Case-Writer
INSUFFICIENT_ARTEFACTS). Avoids the failure mode where the model
refuses honestly but the runtime treated it as a parse error
(observed live with Poirot+Hoover identifying the DIRECTOR
false-positive disambiguation issue in entity_mentions).
Chat tool extensions (web/lib/chat/tools.ts):
- request_investigation now accepts 7 kinds. Each routes to its
detective with appropriate validation (hypothesis_id regex,
person_id kebab-case, topic non-empty, doc_id for evidence_chain).
- ETA per kind: Holmes/Dupin 60s, Poirot 45s, Schneier/Tetlock 30s,
Taleb 50s, Case-Writer 180s (longer prose), Locard 30×n_chunks.
UI integration:
- chat-bubble inline card paints each detective in its tone color.
- /jobs/[id] page header swaps name/subtitle/tone per detective;
question label adapts ("Topic" / "Hypothesis under attack" /
"Witness under analysis" / "Topic to outlier-scan" / "Hypothesis
under recalibration" / "Case to assemble").
- job-status-poller renders: case-report link card (gold), outlier
cards (yellow), witness cards (purple) — alongside existing
hypothesis, evidence, contradiction cards.
- /api/jobs/[id] hydrates witnesses (JOIN entities for canonical_name)
+ gaps (with scope JSONB).
- /c/[slug] page reads /data/ufo/case/reports/<slug>.md and renders
with MarkdownBody, frontmatter parsed for stat chips.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 01:11:39 +00:00
|
|
|
import { MarkdownBody } from "@/components/markdown-body";
|
|
|
|
|
import { AuthBar } from "@/components/auth-bar";
|
W4.1+W4.2: anti-AI-tics house style + bureau nav (back/home everywhere)
Two complaints in one wave:
(W4.1) User: "Não pode ter vícios de IA como uso excessivo de '-' que a IA
coloca geralmente no lugar de vírgulas por exemplo. Isso deve fazer parte
do prompt geral."
- New prompts/_house-style.md banning the 9 most common AI prose tells
in both EN and PT-BR:
1. Em dashes as comma replacements (—)
2. Rule-of-three lists ("concrete, rigorous, and grounded")
3. Conjunctive openers ("Moreover", "Notably", "Ademais")
4. Superficial -ing analyses ("marking a shift", "destacando")
5. Inflated symbolism + AI vocab (tapestry, navigate, delve,
underscore, robust, multifaceted, marco histórico, ...)
6. Negative parallelisms ("Not just X but Y")
7. Vague attribution ("Some scholars say...")
8. Summary closers ("In summary...", "Em suma...")
9. Hedging fluff ("It's important to note...")
Verbatim chunk quotes are explicitly exempt; preserve as-is.
- claude.ts callClaude() lazily loads _house-style.md once per process
and PREPENDS it to every detective's system prompt:
composedSystem = houseStyle + "---" + detective.systemPrompt
This means all 7 detectives + future ones get the rules without any
per-prompt change.
(W4.2) User: "Quando entra em uma página da investigação não tem como
voltar! UX terrível!"
- New <BureauNav> sticky topbar with explicit "← home" + "🔎 bureau"
buttons + clickable breadcrumb trail. Always visible at the top of
every bureau page so the user can escape in one click.
- Wired into /bureau, /h/[hypothesisId], /c/[slug], /jobs/[id]. Each
page passes its sensible parent crumb (/bureau#hypotheses,
/bureau#reports, /bureau#jobs).
- Replaces the previous plain-text "disclosure.top / hypothesis /
H-0004" line which had no visual affordance.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 16:27:58 +00:00
|
|
|
import { BureauNav } from "@/components/bureau-nav";
|
W5.4 (Phase 3B): sitemap + robots + Article schema + magazine reading view
GEO/SEO surface area:
app/robots.ts (new — Next.js dynamic robots)
Explicitly ALLOWS major AI crawlers: GPTBot, OAI-SearchBot,
ChatGPT-User, ClaudeBot, Claude-Web, anthropic-ai, PerplexityBot,
Perplexity-User, Google-Extended, Applebot-Extended, CCBot,
DuckAssistBot, YouBot, Bytespider, Amazonbot. The site exists to
be cited by LLMs answering UAP/UFO questions — we want them in.
/api/admin/, /admin/, /auth/ disallowed for everyone.
app/sitemap.ts (new — Next.js dynamic sitemap)
Lists 9 top-level routes + every /d/<doc> + every /c/<slug> from
the filesystem + up to 500 entity URLs per class
(event, person, uap_object, location, organization),
sorted with summary-enriched entities first. ~3000 URLs total at
current corpus size. lastModified honours summary_generated_at so
crawlers re-index when entities are re-enriched.
app/c/[slug]/page.tsx (rewritten — magazine reading view)
- generateMetadata: per-case title, description (auto-extracted
from the locale-preferred lead paragraph), canonical URL,
hreflang alternate, OpenGraph article type with publishedTime,
Twitter card.
- JSON-LD Article schema embedded at end of page: schema.org
Article + Organization publisher + inLanguage + isAccessibleForFree.
This is what makes the case appear as a citable source in
Google AI Overviews / Perplexity / ChatGPT search.
- Reading view rewritten: display-serif headline (Fraunces), italic
blockquotes with gold accent, prose-typography styling, no more
detective stats line, no more "written by case-writer@detective"
attribution. Locale-aware: PT-BR pulls topic_pt_br + lead in PT,
English mirror.
tailwind.config.ts
+ @tailwindcss/typography plugin
+ font-display family wired to var(--font-display) (Fraunces)
package.json
+ "@tailwindcss/typography" devDependency
Phase 3A note: bulk entity enrichment hit Claude OAuth weekly quota mid-run.
6 events + 3 uap_objects landed bilingual summaries before the quota
exhausted. UI gracefully splits enriched vs bare entities so /sightings
shows the magazine-grade cards (Kenneth Arnold 1947, Roswell, Maury Island,
Joseph Perry 1960 lunar photo, Civil Defense Director 1966, etc.) on top
of a compact table of the rest. Re-run when quota refreshes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 19:09:50 +00:00
|
|
|
import { getLocale } from "@/components/locale-toggle";
|
W3.8: Investigation Bureau complete — Poirot, Taleb, Tetlock, Case-Writer
Brings the bureau from 4 → 8 detectives. All eight run as Bun + claude-CLI
subprocesses against the same Supabase + investigation_jobs LISTEN/NOTIFY
queue, sharing search.ts hybridSearch and writer-side validators that
gate writes against schema + FK.
New detectives:
Poirot (witness_analysis)
- prompts/poirot.md — credibility / access / bias / corroboration /
verdict; uses entity_mentions JOIN chunks to pull 12 chunks per
person; resolves corroboration_refs chunk_ids defensively (accepts
bare cNNNN even when the model emits pNNN/cNNNN).
- INSERT into public.witnesses with W-NNNN naming.
- Tone: purple (#9b5de5).
Taleb (outlier_scan)
- prompts/taleb.md — "surprise is relative to a model"; at most 3
outliers; each requires explicit dominant_model + why_surprising +
what_it_implies; fan-out into public.gaps with scope.kind="outlier".
- Same unscoped-fallback as Dupin (Pass 1 with doc_id, Pass 2 widens
to corpus if hits < 3).
- Tone: yellow (#ffd23f).
Tetlock (calibrate_hypothesis)
- prompts/tetlock.md — honest Bayesian update; emits new_posterior +
Δ + recommended_action ∈ {keep, downgrade, upgrade, supersede}.
- write_calibration UPDATEs public.hypotheses + APPENDS a
"## Calibration history" section to the H-NNNN.md case file
(calibration is append-only — each datapoint matters). Posterior
band auto-corrected to match Tetlock thresholds.
- NO_NEW_EVIDENCE sentinel handled; pure 'keep' with |Δ|<0.005 only
touches updated_at + reviewed_by.
- Tone: teal (#26d4cc).
Case-Writer (case_report)
- prompts/case-writer.md — Dr. Watson assembles all artefacts
(E-NNNN, H-NNNN, R-NNNN, W-NNNN, G-NNNN) into a five-act narrative.
ILIKE filter on topic; doc_id optional scope.
- Larger budget cap (≥ $0.50) + longer timeout for prose generation.
- Writes case/reports/<slug>.md with frontmatter (topic + counts);
no DB table for v0.
- New page /c/[slug] renders the report via MarkdownBody + stat chips.
- Tone: gold (#e0c080).
Hardening across the bureau:
- Sentinel parsing now accepts backticked AND prose-trailing forms
(Holmes NO_HYPOTHESES, Dupin NO_CONTRADICTIONS, Schneier
INSUFFICIENT_HYPOTHESIS, Poirot INSUFFICIENT_TESTIMONY, Taleb
NO_OUTLIERS, Tetlock NO_NEW_EVIDENCE, Case-Writer
INSUFFICIENT_ARTEFACTS). Avoids the failure mode where the model
refuses honestly but the runtime treated it as a parse error
(observed live with Poirot+Hoover identifying the DIRECTOR
false-positive disambiguation issue in entity_mentions).
Chat tool extensions (web/lib/chat/tools.ts):
- request_investigation now accepts 7 kinds. Each routes to its
detective with appropriate validation (hypothesis_id regex,
person_id kebab-case, topic non-empty, doc_id for evidence_chain).
- ETA per kind: Holmes/Dupin 60s, Poirot 45s, Schneier/Tetlock 30s,
Taleb 50s, Case-Writer 180s (longer prose), Locard 30×n_chunks.
UI integration:
- chat-bubble inline card paints each detective in its tone color.
- /jobs/[id] page header swaps name/subtitle/tone per detective;
question label adapts ("Topic" / "Hypothesis under attack" /
"Witness under analysis" / "Topic to outlier-scan" / "Hypothesis
under recalibration" / "Case to assemble").
- job-status-poller renders: case-report link card (gold), outlier
cards (yellow), witness cards (purple) — alongside existing
hypothesis, evidence, contradiction cards.
- /api/jobs/[id] hydrates witnesses (JOIN entities for canonical_name)
+ gaps (with scope JSONB).
- /c/[slug] page reads /data/ufo/case/reports/<slug>.md and renders
with MarkdownBody, frontmatter parsed for stat chips.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 01:11:39 +00:00
|
|
|
|
|
|
|
|
export const runtime = "nodejs";
|
|
|
|
|
export const dynamic = "force-dynamic";
|
|
|
|
|
|
|
|
|
|
const CASE_ROOT = process.env.CASE_ROOT || "/data/ufo/case";
|
W5.4 (Phase 3B): sitemap + robots + Article schema + magazine reading view
GEO/SEO surface area:
app/robots.ts (new — Next.js dynamic robots)
Explicitly ALLOWS major AI crawlers: GPTBot, OAI-SearchBot,
ChatGPT-User, ClaudeBot, Claude-Web, anthropic-ai, PerplexityBot,
Perplexity-User, Google-Extended, Applebot-Extended, CCBot,
DuckAssistBot, YouBot, Bytespider, Amazonbot. The site exists to
be cited by LLMs answering UAP/UFO questions — we want them in.
/api/admin/, /admin/, /auth/ disallowed for everyone.
app/sitemap.ts (new — Next.js dynamic sitemap)
Lists 9 top-level routes + every /d/<doc> + every /c/<slug> from
the filesystem + up to 500 entity URLs per class
(event, person, uap_object, location, organization),
sorted with summary-enriched entities first. ~3000 URLs total at
current corpus size. lastModified honours summary_generated_at so
crawlers re-index when entities are re-enriched.
app/c/[slug]/page.tsx (rewritten — magazine reading view)
- generateMetadata: per-case title, description (auto-extracted
from the locale-preferred lead paragraph), canonical URL,
hreflang alternate, OpenGraph article type with publishedTime,
Twitter card.
- JSON-LD Article schema embedded at end of page: schema.org
Article + Organization publisher + inLanguage + isAccessibleForFree.
This is what makes the case appear as a citable source in
Google AI Overviews / Perplexity / ChatGPT search.
- Reading view rewritten: display-serif headline (Fraunces), italic
blockquotes with gold accent, prose-typography styling, no more
detective stats line, no more "written by case-writer@detective"
attribution. Locale-aware: PT-BR pulls topic_pt_br + lead in PT,
English mirror.
tailwind.config.ts
+ @tailwindcss/typography plugin
+ font-display family wired to var(--font-display) (Fraunces)
package.json
+ "@tailwindcss/typography" devDependency
Phase 3A note: bulk entity enrichment hit Claude OAuth weekly quota mid-run.
6 events + 3 uap_objects landed bilingual summaries before the quota
exhausted. UI gracefully splits enriched vs bare entities so /sightings
shows the magazine-grade cards (Kenneth Arnold 1947, Roswell, Maury Island,
Joseph Perry 1960 lunar photo, Civil Defense Director 1966, etc.) on top
of a compact table of the rest. Re-run when quota refreshes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 19:09:50 +00:00
|
|
|
const SITE_URL = process.env.NEXT_PUBLIC_SITE_URL || "https://disclosure.top";
|
W3.8: Investigation Bureau complete — Poirot, Taleb, Tetlock, Case-Writer
Brings the bureau from 4 → 8 detectives. All eight run as Bun + claude-CLI
subprocesses against the same Supabase + investigation_jobs LISTEN/NOTIFY
queue, sharing search.ts hybridSearch and writer-side validators that
gate writes against schema + FK.
New detectives:
Poirot (witness_analysis)
- prompts/poirot.md — credibility / access / bias / corroboration /
verdict; uses entity_mentions JOIN chunks to pull 12 chunks per
person; resolves corroboration_refs chunk_ids defensively (accepts
bare cNNNN even when the model emits pNNN/cNNNN).
- INSERT into public.witnesses with W-NNNN naming.
- Tone: purple (#9b5de5).
Taleb (outlier_scan)
- prompts/taleb.md — "surprise is relative to a model"; at most 3
outliers; each requires explicit dominant_model + why_surprising +
what_it_implies; fan-out into public.gaps with scope.kind="outlier".
- Same unscoped-fallback as Dupin (Pass 1 with doc_id, Pass 2 widens
to corpus if hits < 3).
- Tone: yellow (#ffd23f).
Tetlock (calibrate_hypothesis)
- prompts/tetlock.md — honest Bayesian update; emits new_posterior +
Δ + recommended_action ∈ {keep, downgrade, upgrade, supersede}.
- write_calibration UPDATEs public.hypotheses + APPENDS a
"## Calibration history" section to the H-NNNN.md case file
(calibration is append-only — each datapoint matters). Posterior
band auto-corrected to match Tetlock thresholds.
- NO_NEW_EVIDENCE sentinel handled; pure 'keep' with |Δ|<0.005 only
touches updated_at + reviewed_by.
- Tone: teal (#26d4cc).
Case-Writer (case_report)
- prompts/case-writer.md — Dr. Watson assembles all artefacts
(E-NNNN, H-NNNN, R-NNNN, W-NNNN, G-NNNN) into a five-act narrative.
ILIKE filter on topic; doc_id optional scope.
- Larger budget cap (≥ $0.50) + longer timeout for prose generation.
- Writes case/reports/<slug>.md with frontmatter (topic + counts);
no DB table for v0.
- New page /c/[slug] renders the report via MarkdownBody + stat chips.
- Tone: gold (#e0c080).
Hardening across the bureau:
- Sentinel parsing now accepts backticked AND prose-trailing forms
(Holmes NO_HYPOTHESES, Dupin NO_CONTRADICTIONS, Schneier
INSUFFICIENT_HYPOTHESIS, Poirot INSUFFICIENT_TESTIMONY, Taleb
NO_OUTLIERS, Tetlock NO_NEW_EVIDENCE, Case-Writer
INSUFFICIENT_ARTEFACTS). Avoids the failure mode where the model
refuses honestly but the runtime treated it as a parse error
(observed live with Poirot+Hoover identifying the DIRECTOR
false-positive disambiguation issue in entity_mentions).
Chat tool extensions (web/lib/chat/tools.ts):
- request_investigation now accepts 7 kinds. Each routes to its
detective with appropriate validation (hypothesis_id regex,
person_id kebab-case, topic non-empty, doc_id for evidence_chain).
- ETA per kind: Holmes/Dupin 60s, Poirot 45s, Schneier/Tetlock 30s,
Taleb 50s, Case-Writer 180s (longer prose), Locard 30×n_chunks.
UI integration:
- chat-bubble inline card paints each detective in its tone color.
- /jobs/[id] page header swaps name/subtitle/tone per detective;
question label adapts ("Topic" / "Hypothesis under attack" /
"Witness under analysis" / "Topic to outlier-scan" / "Hypothesis
under recalibration" / "Case to assemble").
- job-status-poller renders: case-report link card (gold), outlier
cards (yellow), witness cards (purple) — alongside existing
hypothesis, evidence, contradiction cards.
- /api/jobs/[id] hydrates witnesses (JOIN entities for canonical_name)
+ gaps (with scope JSONB).
- /c/[slug] page reads /data/ufo/case/reports/<slug>.md and renders
with MarkdownBody, frontmatter parsed for stat chips.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 01:11:39 +00:00
|
|
|
|
|
|
|
|
interface Frontmatter {
|
|
|
|
|
topic?: string;
|
W5.4 (Phase 3B): sitemap + robots + Article schema + magazine reading view
GEO/SEO surface area:
app/robots.ts (new — Next.js dynamic robots)
Explicitly ALLOWS major AI crawlers: GPTBot, OAI-SearchBot,
ChatGPT-User, ClaudeBot, Claude-Web, anthropic-ai, PerplexityBot,
Perplexity-User, Google-Extended, Applebot-Extended, CCBot,
DuckAssistBot, YouBot, Bytespider, Amazonbot. The site exists to
be cited by LLMs answering UAP/UFO questions — we want them in.
/api/admin/, /admin/, /auth/ disallowed for everyone.
app/sitemap.ts (new — Next.js dynamic sitemap)
Lists 9 top-level routes + every /d/<doc> + every /c/<slug> from
the filesystem + up to 500 entity URLs per class
(event, person, uap_object, location, organization),
sorted with summary-enriched entities first. ~3000 URLs total at
current corpus size. lastModified honours summary_generated_at so
crawlers re-index when entities are re-enriched.
app/c/[slug]/page.tsx (rewritten — magazine reading view)
- generateMetadata: per-case title, description (auto-extracted
from the locale-preferred lead paragraph), canonical URL,
hreflang alternate, OpenGraph article type with publishedTime,
Twitter card.
- JSON-LD Article schema embedded at end of page: schema.org
Article + Organization publisher + inLanguage + isAccessibleForFree.
This is what makes the case appear as a citable source in
Google AI Overviews / Perplexity / ChatGPT search.
- Reading view rewritten: display-serif headline (Fraunces), italic
blockquotes with gold accent, prose-typography styling, no more
detective stats line, no more "written by case-writer@detective"
attribution. Locale-aware: PT-BR pulls topic_pt_br + lead in PT,
English mirror.
tailwind.config.ts
+ @tailwindcss/typography plugin
+ font-display family wired to var(--font-display) (Fraunces)
package.json
+ "@tailwindcss/typography" devDependency
Phase 3A note: bulk entity enrichment hit Claude OAuth weekly quota mid-run.
6 events + 3 uap_objects landed bilingual summaries before the quota
exhausted. UI gracefully splits enriched vs bare entities so /sightings
shows the magazine-grade cards (Kenneth Arnold 1947, Roswell, Maury Island,
Joseph Perry 1960 lunar photo, Civil Defense Director 1966, etc.) on top
of a compact table of the rest. Re-run when quota refreshes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 19:09:50 +00:00
|
|
|
topic_pt_br?: string;
|
W3.8: Investigation Bureau complete — Poirot, Taleb, Tetlock, Case-Writer
Brings the bureau from 4 → 8 detectives. All eight run as Bun + claude-CLI
subprocesses against the same Supabase + investigation_jobs LISTEN/NOTIFY
queue, sharing search.ts hybridSearch and writer-side validators that
gate writes against schema + FK.
New detectives:
Poirot (witness_analysis)
- prompts/poirot.md — credibility / access / bias / corroboration /
verdict; uses entity_mentions JOIN chunks to pull 12 chunks per
person; resolves corroboration_refs chunk_ids defensively (accepts
bare cNNNN even when the model emits pNNN/cNNNN).
- INSERT into public.witnesses with W-NNNN naming.
- Tone: purple (#9b5de5).
Taleb (outlier_scan)
- prompts/taleb.md — "surprise is relative to a model"; at most 3
outliers; each requires explicit dominant_model + why_surprising +
what_it_implies; fan-out into public.gaps with scope.kind="outlier".
- Same unscoped-fallback as Dupin (Pass 1 with doc_id, Pass 2 widens
to corpus if hits < 3).
- Tone: yellow (#ffd23f).
Tetlock (calibrate_hypothesis)
- prompts/tetlock.md — honest Bayesian update; emits new_posterior +
Δ + recommended_action ∈ {keep, downgrade, upgrade, supersede}.
- write_calibration UPDATEs public.hypotheses + APPENDS a
"## Calibration history" section to the H-NNNN.md case file
(calibration is append-only — each datapoint matters). Posterior
band auto-corrected to match Tetlock thresholds.
- NO_NEW_EVIDENCE sentinel handled; pure 'keep' with |Δ|<0.005 only
touches updated_at + reviewed_by.
- Tone: teal (#26d4cc).
Case-Writer (case_report)
- prompts/case-writer.md — Dr. Watson assembles all artefacts
(E-NNNN, H-NNNN, R-NNNN, W-NNNN, G-NNNN) into a five-act narrative.
ILIKE filter on topic; doc_id optional scope.
- Larger budget cap (≥ $0.50) + longer timeout for prose generation.
- Writes case/reports/<slug>.md with frontmatter (topic + counts);
no DB table for v0.
- New page /c/[slug] renders the report via MarkdownBody + stat chips.
- Tone: gold (#e0c080).
Hardening across the bureau:
- Sentinel parsing now accepts backticked AND prose-trailing forms
(Holmes NO_HYPOTHESES, Dupin NO_CONTRADICTIONS, Schneier
INSUFFICIENT_HYPOTHESIS, Poirot INSUFFICIENT_TESTIMONY, Taleb
NO_OUTLIERS, Tetlock NO_NEW_EVIDENCE, Case-Writer
INSUFFICIENT_ARTEFACTS). Avoids the failure mode where the model
refuses honestly but the runtime treated it as a parse error
(observed live with Poirot+Hoover identifying the DIRECTOR
false-positive disambiguation issue in entity_mentions).
Chat tool extensions (web/lib/chat/tools.ts):
- request_investigation now accepts 7 kinds. Each routes to its
detective with appropriate validation (hypothesis_id regex,
person_id kebab-case, topic non-empty, doc_id for evidence_chain).
- ETA per kind: Holmes/Dupin 60s, Poirot 45s, Schneier/Tetlock 30s,
Taleb 50s, Case-Writer 180s (longer prose), Locard 30×n_chunks.
UI integration:
- chat-bubble inline card paints each detective in its tone color.
- /jobs/[id] page header swaps name/subtitle/tone per detective;
question label adapts ("Topic" / "Hypothesis under attack" /
"Witness under analysis" / "Topic to outlier-scan" / "Hypothesis
under recalibration" / "Case to assemble").
- job-status-poller renders: case-report link card (gold), outlier
cards (yellow), witness cards (purple) — alongside existing
hypothesis, evidence, contradiction cards.
- /api/jobs/[id] hydrates witnesses (JOIN entities for canonical_name)
+ gaps (with scope JSONB).
- /c/[slug] page reads /data/ufo/case/reports/<slug>.md and renders
with MarkdownBody, frontmatter parsed for stat chips.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 01:11:39 +00:00
|
|
|
created_at?: string;
|
|
|
|
|
job_id?: string;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
function parseFrontmatter(md: string): { fm: Frontmatter; body: string } {
|
|
|
|
|
const m = md.match(/^---\n([\s\S]+?)\n---\n([\s\S]*)$/);
|
|
|
|
|
if (!m) return { fm: {}, body: md };
|
|
|
|
|
const fm: Frontmatter = {};
|
|
|
|
|
for (const line of m[1].split("\n")) {
|
|
|
|
|
const kv = line.match(/^([a-z_]+):\s*(.+)$/);
|
|
|
|
|
if (!kv) continue;
|
W5.4 (Phase 3B): sitemap + robots + Article schema + magazine reading view
GEO/SEO surface area:
app/robots.ts (new — Next.js dynamic robots)
Explicitly ALLOWS major AI crawlers: GPTBot, OAI-SearchBot,
ChatGPT-User, ClaudeBot, Claude-Web, anthropic-ai, PerplexityBot,
Perplexity-User, Google-Extended, Applebot-Extended, CCBot,
DuckAssistBot, YouBot, Bytespider, Amazonbot. The site exists to
be cited by LLMs answering UAP/UFO questions — we want them in.
/api/admin/, /admin/, /auth/ disallowed for everyone.
app/sitemap.ts (new — Next.js dynamic sitemap)
Lists 9 top-level routes + every /d/<doc> + every /c/<slug> from
the filesystem + up to 500 entity URLs per class
(event, person, uap_object, location, organization),
sorted with summary-enriched entities first. ~3000 URLs total at
current corpus size. lastModified honours summary_generated_at so
crawlers re-index when entities are re-enriched.
app/c/[slug]/page.tsx (rewritten — magazine reading view)
- generateMetadata: per-case title, description (auto-extracted
from the locale-preferred lead paragraph), canonical URL,
hreflang alternate, OpenGraph article type with publishedTime,
Twitter card.
- JSON-LD Article schema embedded at end of page: schema.org
Article + Organization publisher + inLanguage + isAccessibleForFree.
This is what makes the case appear as a citable source in
Google AI Overviews / Perplexity / ChatGPT search.
- Reading view rewritten: display-serif headline (Fraunces), italic
blockquotes with gold accent, prose-typography styling, no more
detective stats line, no more "written by case-writer@detective"
attribution. Locale-aware: PT-BR pulls topic_pt_br + lead in PT,
English mirror.
tailwind.config.ts
+ @tailwindcss/typography plugin
+ font-display family wired to var(--font-display) (Fraunces)
package.json
+ "@tailwindcss/typography" devDependency
Phase 3A note: bulk entity enrichment hit Claude OAuth weekly quota mid-run.
6 events + 3 uap_objects landed bilingual summaries before the quota
exhausted. UI gracefully splits enriched vs bare entities so /sightings
shows the magazine-grade cards (Kenneth Arnold 1947, Roswell, Maury Island,
Joseph Perry 1960 lunar photo, Civil Defense Director 1966, etc.) on top
of a compact table of the rest. Re-run when quota refreshes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 19:09:50 +00:00
|
|
|
let v = kv[2].trim();
|
W3.8: Investigation Bureau complete — Poirot, Taleb, Tetlock, Case-Writer
Brings the bureau from 4 → 8 detectives. All eight run as Bun + claude-CLI
subprocesses against the same Supabase + investigation_jobs LISTEN/NOTIFY
queue, sharing search.ts hybridSearch and writer-side validators that
gate writes against schema + FK.
New detectives:
Poirot (witness_analysis)
- prompts/poirot.md — credibility / access / bias / corroboration /
verdict; uses entity_mentions JOIN chunks to pull 12 chunks per
person; resolves corroboration_refs chunk_ids defensively (accepts
bare cNNNN even when the model emits pNNN/cNNNN).
- INSERT into public.witnesses with W-NNNN naming.
- Tone: purple (#9b5de5).
Taleb (outlier_scan)
- prompts/taleb.md — "surprise is relative to a model"; at most 3
outliers; each requires explicit dominant_model + why_surprising +
what_it_implies; fan-out into public.gaps with scope.kind="outlier".
- Same unscoped-fallback as Dupin (Pass 1 with doc_id, Pass 2 widens
to corpus if hits < 3).
- Tone: yellow (#ffd23f).
Tetlock (calibrate_hypothesis)
- prompts/tetlock.md — honest Bayesian update; emits new_posterior +
Δ + recommended_action ∈ {keep, downgrade, upgrade, supersede}.
- write_calibration UPDATEs public.hypotheses + APPENDS a
"## Calibration history" section to the H-NNNN.md case file
(calibration is append-only — each datapoint matters). Posterior
band auto-corrected to match Tetlock thresholds.
- NO_NEW_EVIDENCE sentinel handled; pure 'keep' with |Δ|<0.005 only
touches updated_at + reviewed_by.
- Tone: teal (#26d4cc).
Case-Writer (case_report)
- prompts/case-writer.md — Dr. Watson assembles all artefacts
(E-NNNN, H-NNNN, R-NNNN, W-NNNN, G-NNNN) into a five-act narrative.
ILIKE filter on topic; doc_id optional scope.
- Larger budget cap (≥ $0.50) + longer timeout for prose generation.
- Writes case/reports/<slug>.md with frontmatter (topic + counts);
no DB table for v0.
- New page /c/[slug] renders the report via MarkdownBody + stat chips.
- Tone: gold (#e0c080).
Hardening across the bureau:
- Sentinel parsing now accepts backticked AND prose-trailing forms
(Holmes NO_HYPOTHESES, Dupin NO_CONTRADICTIONS, Schneier
INSUFFICIENT_HYPOTHESIS, Poirot INSUFFICIENT_TESTIMONY, Taleb
NO_OUTLIERS, Tetlock NO_NEW_EVIDENCE, Case-Writer
INSUFFICIENT_ARTEFACTS). Avoids the failure mode where the model
refuses honestly but the runtime treated it as a parse error
(observed live with Poirot+Hoover identifying the DIRECTOR
false-positive disambiguation issue in entity_mentions).
Chat tool extensions (web/lib/chat/tools.ts):
- request_investigation now accepts 7 kinds. Each routes to its
detective with appropriate validation (hypothesis_id regex,
person_id kebab-case, topic non-empty, doc_id for evidence_chain).
- ETA per kind: Holmes/Dupin 60s, Poirot 45s, Schneier/Tetlock 30s,
Taleb 50s, Case-Writer 180s (longer prose), Locard 30×n_chunks.
UI integration:
- chat-bubble inline card paints each detective in its tone color.
- /jobs/[id] page header swaps name/subtitle/tone per detective;
question label adapts ("Topic" / "Hypothesis under attack" /
"Witness under analysis" / "Topic to outlier-scan" / "Hypothesis
under recalibration" / "Case to assemble").
- job-status-poller renders: case-report link card (gold), outlier
cards (yellow), witness cards (purple) — alongside existing
hypothesis, evidence, contradiction cards.
- /api/jobs/[id] hydrates witnesses (JOIN entities for canonical_name)
+ gaps (with scope JSONB).
- /c/[slug] page reads /data/ufo/case/reports/<slug>.md and renders
with MarkdownBody, frontmatter parsed for stat chips.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 01:11:39 +00:00
|
|
|
if (v.startsWith('"') && v.endsWith('"')) v = v.slice(1, -1);
|
W5.4 (Phase 3B): sitemap + robots + Article schema + magazine reading view
GEO/SEO surface area:
app/robots.ts (new — Next.js dynamic robots)
Explicitly ALLOWS major AI crawlers: GPTBot, OAI-SearchBot,
ChatGPT-User, ClaudeBot, Claude-Web, anthropic-ai, PerplexityBot,
Perplexity-User, Google-Extended, Applebot-Extended, CCBot,
DuckAssistBot, YouBot, Bytespider, Amazonbot. The site exists to
be cited by LLMs answering UAP/UFO questions — we want them in.
/api/admin/, /admin/, /auth/ disallowed for everyone.
app/sitemap.ts (new — Next.js dynamic sitemap)
Lists 9 top-level routes + every /d/<doc> + every /c/<slug> from
the filesystem + up to 500 entity URLs per class
(event, person, uap_object, location, organization),
sorted with summary-enriched entities first. ~3000 URLs total at
current corpus size. lastModified honours summary_generated_at so
crawlers re-index when entities are re-enriched.
app/c/[slug]/page.tsx (rewritten — magazine reading view)
- generateMetadata: per-case title, description (auto-extracted
from the locale-preferred lead paragraph), canonical URL,
hreflang alternate, OpenGraph article type with publishedTime,
Twitter card.
- JSON-LD Article schema embedded at end of page: schema.org
Article + Organization publisher + inLanguage + isAccessibleForFree.
This is what makes the case appear as a citable source in
Google AI Overviews / Perplexity / ChatGPT search.
- Reading view rewritten: display-serif headline (Fraunces), italic
blockquotes with gold accent, prose-typography styling, no more
detective stats line, no more "written by case-writer@detective"
attribution. Locale-aware: PT-BR pulls topic_pt_br + lead in PT,
English mirror.
tailwind.config.ts
+ @tailwindcss/typography plugin
+ font-display family wired to var(--font-display) (Fraunces)
package.json
+ "@tailwindcss/typography" devDependency
Phase 3A note: bulk entity enrichment hit Claude OAuth weekly quota mid-run.
6 events + 3 uap_objects landed bilingual summaries before the quota
exhausted. UI gracefully splits enriched vs bare entities so /sightings
shows the magazine-grade cards (Kenneth Arnold 1947, Roswell, Maury Island,
Joseph Perry 1960 lunar photo, Civil Defense Director 1966, etc.) on top
of a compact table of the rest. Re-run when quota refreshes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 19:09:50 +00:00
|
|
|
(fm as Record<string, string>)[kv[1]] = v;
|
W3.8: Investigation Bureau complete — Poirot, Taleb, Tetlock, Case-Writer
Brings the bureau from 4 → 8 detectives. All eight run as Bun + claude-CLI
subprocesses against the same Supabase + investigation_jobs LISTEN/NOTIFY
queue, sharing search.ts hybridSearch and writer-side validators that
gate writes against schema + FK.
New detectives:
Poirot (witness_analysis)
- prompts/poirot.md — credibility / access / bias / corroboration /
verdict; uses entity_mentions JOIN chunks to pull 12 chunks per
person; resolves corroboration_refs chunk_ids defensively (accepts
bare cNNNN even when the model emits pNNN/cNNNN).
- INSERT into public.witnesses with W-NNNN naming.
- Tone: purple (#9b5de5).
Taleb (outlier_scan)
- prompts/taleb.md — "surprise is relative to a model"; at most 3
outliers; each requires explicit dominant_model + why_surprising +
what_it_implies; fan-out into public.gaps with scope.kind="outlier".
- Same unscoped-fallback as Dupin (Pass 1 with doc_id, Pass 2 widens
to corpus if hits < 3).
- Tone: yellow (#ffd23f).
Tetlock (calibrate_hypothesis)
- prompts/tetlock.md — honest Bayesian update; emits new_posterior +
Δ + recommended_action ∈ {keep, downgrade, upgrade, supersede}.
- write_calibration UPDATEs public.hypotheses + APPENDS a
"## Calibration history" section to the H-NNNN.md case file
(calibration is append-only — each datapoint matters). Posterior
band auto-corrected to match Tetlock thresholds.
- NO_NEW_EVIDENCE sentinel handled; pure 'keep' with |Δ|<0.005 only
touches updated_at + reviewed_by.
- Tone: teal (#26d4cc).
Case-Writer (case_report)
- prompts/case-writer.md — Dr. Watson assembles all artefacts
(E-NNNN, H-NNNN, R-NNNN, W-NNNN, G-NNNN) into a five-act narrative.
ILIKE filter on topic; doc_id optional scope.
- Larger budget cap (≥ $0.50) + longer timeout for prose generation.
- Writes case/reports/<slug>.md with frontmatter (topic + counts);
no DB table for v0.
- New page /c/[slug] renders the report via MarkdownBody + stat chips.
- Tone: gold (#e0c080).
Hardening across the bureau:
- Sentinel parsing now accepts backticked AND prose-trailing forms
(Holmes NO_HYPOTHESES, Dupin NO_CONTRADICTIONS, Schneier
INSUFFICIENT_HYPOTHESIS, Poirot INSUFFICIENT_TESTIMONY, Taleb
NO_OUTLIERS, Tetlock NO_NEW_EVIDENCE, Case-Writer
INSUFFICIENT_ARTEFACTS). Avoids the failure mode where the model
refuses honestly but the runtime treated it as a parse error
(observed live with Poirot+Hoover identifying the DIRECTOR
false-positive disambiguation issue in entity_mentions).
Chat tool extensions (web/lib/chat/tools.ts):
- request_investigation now accepts 7 kinds. Each routes to its
detective with appropriate validation (hypothesis_id regex,
person_id kebab-case, topic non-empty, doc_id for evidence_chain).
- ETA per kind: Holmes/Dupin 60s, Poirot 45s, Schneier/Tetlock 30s,
Taleb 50s, Case-Writer 180s (longer prose), Locard 30×n_chunks.
UI integration:
- chat-bubble inline card paints each detective in its tone color.
- /jobs/[id] page header swaps name/subtitle/tone per detective;
question label adapts ("Topic" / "Hypothesis under attack" /
"Witness under analysis" / "Topic to outlier-scan" / "Hypothesis
under recalibration" / "Case to assemble").
- job-status-poller renders: case-report link card (gold), outlier
cards (yellow), witness cards (purple) — alongside existing
hypothesis, evidence, contradiction cards.
- /api/jobs/[id] hydrates witnesses (JOIN entities for canonical_name)
+ gaps (with scope JSONB).
- /c/[slug] page reads /data/ufo/case/reports/<slug>.md and renders
with MarkdownBody, frontmatter parsed for stat chips.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 01:11:39 +00:00
|
|
|
}
|
|
|
|
|
return { fm, body: m[2] };
|
|
|
|
|
}
|
|
|
|
|
|
W5.4 (Phase 3B): sitemap + robots + Article schema + magazine reading view
GEO/SEO surface area:
app/robots.ts (new — Next.js dynamic robots)
Explicitly ALLOWS major AI crawlers: GPTBot, OAI-SearchBot,
ChatGPT-User, ClaudeBot, Claude-Web, anthropic-ai, PerplexityBot,
Perplexity-User, Google-Extended, Applebot-Extended, CCBot,
DuckAssistBot, YouBot, Bytespider, Amazonbot. The site exists to
be cited by LLMs answering UAP/UFO questions — we want them in.
/api/admin/, /admin/, /auth/ disallowed for everyone.
app/sitemap.ts (new — Next.js dynamic sitemap)
Lists 9 top-level routes + every /d/<doc> + every /c/<slug> from
the filesystem + up to 500 entity URLs per class
(event, person, uap_object, location, organization),
sorted with summary-enriched entities first. ~3000 URLs total at
current corpus size. lastModified honours summary_generated_at so
crawlers re-index when entities are re-enriched.
app/c/[slug]/page.tsx (rewritten — magazine reading view)
- generateMetadata: per-case title, description (auto-extracted
from the locale-preferred lead paragraph), canonical URL,
hreflang alternate, OpenGraph article type with publishedTime,
Twitter card.
- JSON-LD Article schema embedded at end of page: schema.org
Article + Organization publisher + inLanguage + isAccessibleForFree.
This is what makes the case appear as a citable source in
Google AI Overviews / Perplexity / ChatGPT search.
- Reading view rewritten: display-serif headline (Fraunces), italic
blockquotes with gold accent, prose-typography styling, no more
detective stats line, no more "written by case-writer@detective"
attribution. Locale-aware: PT-BR pulls topic_pt_br + lead in PT,
English mirror.
tailwind.config.ts
+ @tailwindcss/typography plugin
+ font-display family wired to var(--font-display) (Fraunces)
package.json
+ "@tailwindcss/typography" devDependency
Phase 3A note: bulk entity enrichment hit Claude OAuth weekly quota mid-run.
6 events + 3 uap_objects landed bilingual summaries before the quota
exhausted. UI gracefully splits enriched vs bare entities so /sightings
shows the magazine-grade cards (Kenneth Arnold 1947, Roswell, Maury Island,
Joseph Perry 1960 lunar photo, Civil Defense Director 1966, etc.) on top
of a compact table of the rest. Re-run when quota refreshes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 19:09:50 +00:00
|
|
|
async function loadCase(slug: string): Promise<{ fm: Frontmatter; body: string } | null> {
|
|
|
|
|
try {
|
|
|
|
|
const md = await readFile(path.join(CASE_ROOT, "reports", `${slug}.md`), "utf-8");
|
|
|
|
|
return parseFrontmatter(md);
|
|
|
|
|
} catch {
|
|
|
|
|
return null;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
2026-05-24 19:16:20 +00:00
|
|
|
async function hasIllustration(slug: string): Promise<boolean> {
|
|
|
|
|
const fs = await import("node:fs/promises");
|
|
|
|
|
const UFO_ROOT = process.env.UFO_ROOT || "/data/ufo";
|
|
|
|
|
try {
|
|
|
|
|
await fs.stat(path.join(UFO_ROOT, "processing", "case-art", `${slug}.png`));
|
|
|
|
|
return true;
|
|
|
|
|
} catch {
|
|
|
|
|
return false;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
W5.4 (Phase 3B): sitemap + robots + Article schema + magazine reading view
GEO/SEO surface area:
app/robots.ts (new — Next.js dynamic robots)
Explicitly ALLOWS major AI crawlers: GPTBot, OAI-SearchBot,
ChatGPT-User, ClaudeBot, Claude-Web, anthropic-ai, PerplexityBot,
Perplexity-User, Google-Extended, Applebot-Extended, CCBot,
DuckAssistBot, YouBot, Bytespider, Amazonbot. The site exists to
be cited by LLMs answering UAP/UFO questions — we want them in.
/api/admin/, /admin/, /auth/ disallowed for everyone.
app/sitemap.ts (new — Next.js dynamic sitemap)
Lists 9 top-level routes + every /d/<doc> + every /c/<slug> from
the filesystem + up to 500 entity URLs per class
(event, person, uap_object, location, organization),
sorted with summary-enriched entities first. ~3000 URLs total at
current corpus size. lastModified honours summary_generated_at so
crawlers re-index when entities are re-enriched.
app/c/[slug]/page.tsx (rewritten — magazine reading view)
- generateMetadata: per-case title, description (auto-extracted
from the locale-preferred lead paragraph), canonical URL,
hreflang alternate, OpenGraph article type with publishedTime,
Twitter card.
- JSON-LD Article schema embedded at end of page: schema.org
Article + Organization publisher + inLanguage + isAccessibleForFree.
This is what makes the case appear as a citable source in
Google AI Overviews / Perplexity / ChatGPT search.
- Reading view rewritten: display-serif headline (Fraunces), italic
blockquotes with gold accent, prose-typography styling, no more
detective stats line, no more "written by case-writer@detective"
attribution. Locale-aware: PT-BR pulls topic_pt_br + lead in PT,
English mirror.
tailwind.config.ts
+ @tailwindcss/typography plugin
+ font-display family wired to var(--font-display) (Fraunces)
package.json
+ "@tailwindcss/typography" devDependency
Phase 3A note: bulk entity enrichment hit Claude OAuth weekly quota mid-run.
6 events + 3 uap_objects landed bilingual summaries before the quota
exhausted. UI gracefully splits enriched vs bare entities so /sightings
shows the magazine-grade cards (Kenneth Arnold 1947, Roswell, Maury Island,
Joseph Perry 1960 lunar photo, Civil Defense Director 1966, etc.) on top
of a compact table of the rest. Re-run when quota refreshes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 19:09:50 +00:00
|
|
|
/**
|
|
|
|
|
* Extract the first prose paragraph from the body for the meta description
|
|
|
|
|
* + OG. We pick the locale-preferred sub-section's opener.
|
|
|
|
|
*/
|
|
|
|
|
function pickLead(body: string, locale: "pt-br" | "en"): string {
|
|
|
|
|
const marker = locale === "pt-br" ? "(PT-BR)" : "(EN)";
|
|
|
|
|
const idx = body.indexOf(marker);
|
|
|
|
|
const slice = idx >= 0 ? body.slice(idx) : body;
|
|
|
|
|
const m = slice.match(/\n\n([^\n#>|`-][^\n]+(?:\n[^\n#>|`-][^\n]+)*)/);
|
|
|
|
|
return (m?.[1] ?? "").replace(/\s+/g, " ").trim();
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
export async function generateMetadata(
|
|
|
|
|
{ params }: { params: Promise<{ slug: string }> },
|
|
|
|
|
): Promise<Metadata> {
|
|
|
|
|
const { slug } = await params;
|
|
|
|
|
const locale = (await getLocale()) === "en" ? "en" : "pt-br";
|
|
|
|
|
const c = await loadCase(slug);
|
|
|
|
|
if (!c) return { title: "Case file not found" };
|
|
|
|
|
|
|
|
|
|
const title = locale === "pt-br" ? (c.fm.topic_pt_br ?? c.fm.topic ?? slug) : (c.fm.topic ?? slug);
|
|
|
|
|
const desc = pickLead(c.body, locale).slice(0, 200);
|
|
|
|
|
const canonical = `${SITE_URL}/c/${slug}`;
|
2026-05-24 19:55:35 +00:00
|
|
|
// OG image — use the case's editorial illustration when present. WhatsApp,
|
|
|
|
|
// Twitter, Slack, Telegram, ChatGPT search all pull this as the link card.
|
|
|
|
|
const ogImage = (await hasIllustration(slug))
|
|
|
|
|
? `${SITE_URL}/api/static/processing/case-art/${slug}.png`
|
|
|
|
|
: `${SITE_URL}/api/static/processing/case-art/green-fireballs-narrative.png`;
|
W5.4 (Phase 3B): sitemap + robots + Article schema + magazine reading view
GEO/SEO surface area:
app/robots.ts (new — Next.js dynamic robots)
Explicitly ALLOWS major AI crawlers: GPTBot, OAI-SearchBot,
ChatGPT-User, ClaudeBot, Claude-Web, anthropic-ai, PerplexityBot,
Perplexity-User, Google-Extended, Applebot-Extended, CCBot,
DuckAssistBot, YouBot, Bytespider, Amazonbot. The site exists to
be cited by LLMs answering UAP/UFO questions — we want them in.
/api/admin/, /admin/, /auth/ disallowed for everyone.
app/sitemap.ts (new — Next.js dynamic sitemap)
Lists 9 top-level routes + every /d/<doc> + every /c/<slug> from
the filesystem + up to 500 entity URLs per class
(event, person, uap_object, location, organization),
sorted with summary-enriched entities first. ~3000 URLs total at
current corpus size. lastModified honours summary_generated_at so
crawlers re-index when entities are re-enriched.
app/c/[slug]/page.tsx (rewritten — magazine reading view)
- generateMetadata: per-case title, description (auto-extracted
from the locale-preferred lead paragraph), canonical URL,
hreflang alternate, OpenGraph article type with publishedTime,
Twitter card.
- JSON-LD Article schema embedded at end of page: schema.org
Article + Organization publisher + inLanguage + isAccessibleForFree.
This is what makes the case appear as a citable source in
Google AI Overviews / Perplexity / ChatGPT search.
- Reading view rewritten: display-serif headline (Fraunces), italic
blockquotes with gold accent, prose-typography styling, no more
detective stats line, no more "written by case-writer@detective"
attribution. Locale-aware: PT-BR pulls topic_pt_br + lead in PT,
English mirror.
tailwind.config.ts
+ @tailwindcss/typography plugin
+ font-display family wired to var(--font-display) (Fraunces)
package.json
+ "@tailwindcss/typography" devDependency
Phase 3A note: bulk entity enrichment hit Claude OAuth weekly quota mid-run.
6 events + 3 uap_objects landed bilingual summaries before the quota
exhausted. UI gracefully splits enriched vs bare entities so /sightings
shows the magazine-grade cards (Kenneth Arnold 1947, Roswell, Maury Island,
Joseph Perry 1960 lunar photo, Civil Defense Director 1966, etc.) on top
of a compact table of the rest. Re-run when quota refreshes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 19:09:50 +00:00
|
|
|
return {
|
|
|
|
|
title,
|
|
|
|
|
description: desc,
|
|
|
|
|
alternates: { canonical, languages: { "pt-BR": canonical, "en-US": canonical } },
|
|
|
|
|
openGraph: {
|
|
|
|
|
type: "article",
|
|
|
|
|
title,
|
|
|
|
|
description: desc,
|
|
|
|
|
url: canonical,
|
|
|
|
|
siteName: "The Disclosure Bureau",
|
|
|
|
|
locale: locale === "pt-br" ? "pt_BR" : "en_US",
|
|
|
|
|
publishedTime: c.fm.created_at,
|
2026-05-24 19:55:35 +00:00
|
|
|
images: [{ url: ogImage, width: 2000, height: 1125, alt: title }],
|
W5.4 (Phase 3B): sitemap + robots + Article schema + magazine reading view
GEO/SEO surface area:
app/robots.ts (new — Next.js dynamic robots)
Explicitly ALLOWS major AI crawlers: GPTBot, OAI-SearchBot,
ChatGPT-User, ClaudeBot, Claude-Web, anthropic-ai, PerplexityBot,
Perplexity-User, Google-Extended, Applebot-Extended, CCBot,
DuckAssistBot, YouBot, Bytespider, Amazonbot. The site exists to
be cited by LLMs answering UAP/UFO questions — we want them in.
/api/admin/, /admin/, /auth/ disallowed for everyone.
app/sitemap.ts (new — Next.js dynamic sitemap)
Lists 9 top-level routes + every /d/<doc> + every /c/<slug> from
the filesystem + up to 500 entity URLs per class
(event, person, uap_object, location, organization),
sorted with summary-enriched entities first. ~3000 URLs total at
current corpus size. lastModified honours summary_generated_at so
crawlers re-index when entities are re-enriched.
app/c/[slug]/page.tsx (rewritten — magazine reading view)
- generateMetadata: per-case title, description (auto-extracted
from the locale-preferred lead paragraph), canonical URL,
hreflang alternate, OpenGraph article type with publishedTime,
Twitter card.
- JSON-LD Article schema embedded at end of page: schema.org
Article + Organization publisher + inLanguage + isAccessibleForFree.
This is what makes the case appear as a citable source in
Google AI Overviews / Perplexity / ChatGPT search.
- Reading view rewritten: display-serif headline (Fraunces), italic
blockquotes with gold accent, prose-typography styling, no more
detective stats line, no more "written by case-writer@detective"
attribution. Locale-aware: PT-BR pulls topic_pt_br + lead in PT,
English mirror.
tailwind.config.ts
+ @tailwindcss/typography plugin
+ font-display family wired to var(--font-display) (Fraunces)
package.json
+ "@tailwindcss/typography" devDependency
Phase 3A note: bulk entity enrichment hit Claude OAuth weekly quota mid-run.
6 events + 3 uap_objects landed bilingual summaries before the quota
exhausted. UI gracefully splits enriched vs bare entities so /sightings
shows the magazine-grade cards (Kenneth Arnold 1947, Roswell, Maury Island,
Joseph Perry 1960 lunar photo, Civil Defense Director 1966, etc.) on top
of a compact table of the rest. Re-run when quota refreshes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 19:09:50 +00:00
|
|
|
},
|
|
|
|
|
twitter: {
|
|
|
|
|
card: "summary_large_image",
|
|
|
|
|
title,
|
|
|
|
|
description: desc,
|
2026-05-24 19:55:35 +00:00
|
|
|
images: [ogImage],
|
W5.4 (Phase 3B): sitemap + robots + Article schema + magazine reading view
GEO/SEO surface area:
app/robots.ts (new — Next.js dynamic robots)
Explicitly ALLOWS major AI crawlers: GPTBot, OAI-SearchBot,
ChatGPT-User, ClaudeBot, Claude-Web, anthropic-ai, PerplexityBot,
Perplexity-User, Google-Extended, Applebot-Extended, CCBot,
DuckAssistBot, YouBot, Bytespider, Amazonbot. The site exists to
be cited by LLMs answering UAP/UFO questions — we want them in.
/api/admin/, /admin/, /auth/ disallowed for everyone.
app/sitemap.ts (new — Next.js dynamic sitemap)
Lists 9 top-level routes + every /d/<doc> + every /c/<slug> from
the filesystem + up to 500 entity URLs per class
(event, person, uap_object, location, organization),
sorted with summary-enriched entities first. ~3000 URLs total at
current corpus size. lastModified honours summary_generated_at so
crawlers re-index when entities are re-enriched.
app/c/[slug]/page.tsx (rewritten — magazine reading view)
- generateMetadata: per-case title, description (auto-extracted
from the locale-preferred lead paragraph), canonical URL,
hreflang alternate, OpenGraph article type with publishedTime,
Twitter card.
- JSON-LD Article schema embedded at end of page: schema.org
Article + Organization publisher + inLanguage + isAccessibleForFree.
This is what makes the case appear as a citable source in
Google AI Overviews / Perplexity / ChatGPT search.
- Reading view rewritten: display-serif headline (Fraunces), italic
blockquotes with gold accent, prose-typography styling, no more
detective stats line, no more "written by case-writer@detective"
attribution. Locale-aware: PT-BR pulls topic_pt_br + lead in PT,
English mirror.
tailwind.config.ts
+ @tailwindcss/typography plugin
+ font-display family wired to var(--font-display) (Fraunces)
package.json
+ "@tailwindcss/typography" devDependency
Phase 3A note: bulk entity enrichment hit Claude OAuth weekly quota mid-run.
6 events + 3 uap_objects landed bilingual summaries before the quota
exhausted. UI gracefully splits enriched vs bare entities so /sightings
shows the magazine-grade cards (Kenneth Arnold 1947, Roswell, Maury Island,
Joseph Perry 1960 lunar photo, Civil Defense Director 1966, etc.) on top
of a compact table of the rest. Re-run when quota refreshes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 19:09:50 +00:00
|
|
|
},
|
|
|
|
|
};
|
|
|
|
|
}
|
|
|
|
|
|
W3.8: Investigation Bureau complete — Poirot, Taleb, Tetlock, Case-Writer
Brings the bureau from 4 → 8 detectives. All eight run as Bun + claude-CLI
subprocesses against the same Supabase + investigation_jobs LISTEN/NOTIFY
queue, sharing search.ts hybridSearch and writer-side validators that
gate writes against schema + FK.
New detectives:
Poirot (witness_analysis)
- prompts/poirot.md — credibility / access / bias / corroboration /
verdict; uses entity_mentions JOIN chunks to pull 12 chunks per
person; resolves corroboration_refs chunk_ids defensively (accepts
bare cNNNN even when the model emits pNNN/cNNNN).
- INSERT into public.witnesses with W-NNNN naming.
- Tone: purple (#9b5de5).
Taleb (outlier_scan)
- prompts/taleb.md — "surprise is relative to a model"; at most 3
outliers; each requires explicit dominant_model + why_surprising +
what_it_implies; fan-out into public.gaps with scope.kind="outlier".
- Same unscoped-fallback as Dupin (Pass 1 with doc_id, Pass 2 widens
to corpus if hits < 3).
- Tone: yellow (#ffd23f).
Tetlock (calibrate_hypothesis)
- prompts/tetlock.md — honest Bayesian update; emits new_posterior +
Δ + recommended_action ∈ {keep, downgrade, upgrade, supersede}.
- write_calibration UPDATEs public.hypotheses + APPENDS a
"## Calibration history" section to the H-NNNN.md case file
(calibration is append-only — each datapoint matters). Posterior
band auto-corrected to match Tetlock thresholds.
- NO_NEW_EVIDENCE sentinel handled; pure 'keep' with |Δ|<0.005 only
touches updated_at + reviewed_by.
- Tone: teal (#26d4cc).
Case-Writer (case_report)
- prompts/case-writer.md — Dr. Watson assembles all artefacts
(E-NNNN, H-NNNN, R-NNNN, W-NNNN, G-NNNN) into a five-act narrative.
ILIKE filter on topic; doc_id optional scope.
- Larger budget cap (≥ $0.50) + longer timeout for prose generation.
- Writes case/reports/<slug>.md with frontmatter (topic + counts);
no DB table for v0.
- New page /c/[slug] renders the report via MarkdownBody + stat chips.
- Tone: gold (#e0c080).
Hardening across the bureau:
- Sentinel parsing now accepts backticked AND prose-trailing forms
(Holmes NO_HYPOTHESES, Dupin NO_CONTRADICTIONS, Schneier
INSUFFICIENT_HYPOTHESIS, Poirot INSUFFICIENT_TESTIMONY, Taleb
NO_OUTLIERS, Tetlock NO_NEW_EVIDENCE, Case-Writer
INSUFFICIENT_ARTEFACTS). Avoids the failure mode where the model
refuses honestly but the runtime treated it as a parse error
(observed live with Poirot+Hoover identifying the DIRECTOR
false-positive disambiguation issue in entity_mentions).
Chat tool extensions (web/lib/chat/tools.ts):
- request_investigation now accepts 7 kinds. Each routes to its
detective with appropriate validation (hypothesis_id regex,
person_id kebab-case, topic non-empty, doc_id for evidence_chain).
- ETA per kind: Holmes/Dupin 60s, Poirot 45s, Schneier/Tetlock 30s,
Taleb 50s, Case-Writer 180s (longer prose), Locard 30×n_chunks.
UI integration:
- chat-bubble inline card paints each detective in its tone color.
- /jobs/[id] page header swaps name/subtitle/tone per detective;
question label adapts ("Topic" / "Hypothesis under attack" /
"Witness under analysis" / "Topic to outlier-scan" / "Hypothesis
under recalibration" / "Case to assemble").
- job-status-poller renders: case-report link card (gold), outlier
cards (yellow), witness cards (purple) — alongside existing
hypothesis, evidence, contradiction cards.
- /api/jobs/[id] hydrates witnesses (JOIN entities for canonical_name)
+ gaps (with scope JSONB).
- /c/[slug] page reads /data/ufo/case/reports/<slug>.md and renders
with MarkdownBody, frontmatter parsed for stat chips.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 01:11:39 +00:00
|
|
|
export default async function CaseReportPage({
|
|
|
|
|
params,
|
|
|
|
|
}: { params: Promise<{ slug: string }> }) {
|
|
|
|
|
const { slug } = await params;
|
|
|
|
|
if (!/^[a-z0-9][a-z0-9-]*$/.test(slug)) notFound();
|
|
|
|
|
|
W5.4 (Phase 3B): sitemap + robots + Article schema + magazine reading view
GEO/SEO surface area:
app/robots.ts (new — Next.js dynamic robots)
Explicitly ALLOWS major AI crawlers: GPTBot, OAI-SearchBot,
ChatGPT-User, ClaudeBot, Claude-Web, anthropic-ai, PerplexityBot,
Perplexity-User, Google-Extended, Applebot-Extended, CCBot,
DuckAssistBot, YouBot, Bytespider, Amazonbot. The site exists to
be cited by LLMs answering UAP/UFO questions — we want them in.
/api/admin/, /admin/, /auth/ disallowed for everyone.
app/sitemap.ts (new — Next.js dynamic sitemap)
Lists 9 top-level routes + every /d/<doc> + every /c/<slug> from
the filesystem + up to 500 entity URLs per class
(event, person, uap_object, location, organization),
sorted with summary-enriched entities first. ~3000 URLs total at
current corpus size. lastModified honours summary_generated_at so
crawlers re-index when entities are re-enriched.
app/c/[slug]/page.tsx (rewritten — magazine reading view)
- generateMetadata: per-case title, description (auto-extracted
from the locale-preferred lead paragraph), canonical URL,
hreflang alternate, OpenGraph article type with publishedTime,
Twitter card.
- JSON-LD Article schema embedded at end of page: schema.org
Article + Organization publisher + inLanguage + isAccessibleForFree.
This is what makes the case appear as a citable source in
Google AI Overviews / Perplexity / ChatGPT search.
- Reading view rewritten: display-serif headline (Fraunces), italic
blockquotes with gold accent, prose-typography styling, no more
detective stats line, no more "written by case-writer@detective"
attribution. Locale-aware: PT-BR pulls topic_pt_br + lead in PT,
English mirror.
tailwind.config.ts
+ @tailwindcss/typography plugin
+ font-display family wired to var(--font-display) (Fraunces)
package.json
+ "@tailwindcss/typography" devDependency
Phase 3A note: bulk entity enrichment hit Claude OAuth weekly quota mid-run.
6 events + 3 uap_objects landed bilingual summaries before the quota
exhausted. UI gracefully splits enriched vs bare entities so /sightings
shows the magazine-grade cards (Kenneth Arnold 1947, Roswell, Maury Island,
Joseph Perry 1960 lunar photo, Civil Defense Director 1966, etc.) on top
of a compact table of the rest. Re-run when quota refreshes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 19:09:50 +00:00
|
|
|
const locale = (await getLocale()) === "en" ? "en" : "pt-br";
|
|
|
|
|
const c = await loadCase(slug);
|
|
|
|
|
if (!c) notFound();
|
|
|
|
|
const { fm, body } = c;
|
W3.8: Investigation Bureau complete — Poirot, Taleb, Tetlock, Case-Writer
Brings the bureau from 4 → 8 detectives. All eight run as Bun + claude-CLI
subprocesses against the same Supabase + investigation_jobs LISTEN/NOTIFY
queue, sharing search.ts hybridSearch and writer-side validators that
gate writes against schema + FK.
New detectives:
Poirot (witness_analysis)
- prompts/poirot.md — credibility / access / bias / corroboration /
verdict; uses entity_mentions JOIN chunks to pull 12 chunks per
person; resolves corroboration_refs chunk_ids defensively (accepts
bare cNNNN even when the model emits pNNN/cNNNN).
- INSERT into public.witnesses with W-NNNN naming.
- Tone: purple (#9b5de5).
Taleb (outlier_scan)
- prompts/taleb.md — "surprise is relative to a model"; at most 3
outliers; each requires explicit dominant_model + why_surprising +
what_it_implies; fan-out into public.gaps with scope.kind="outlier".
- Same unscoped-fallback as Dupin (Pass 1 with doc_id, Pass 2 widens
to corpus if hits < 3).
- Tone: yellow (#ffd23f).
Tetlock (calibrate_hypothesis)
- prompts/tetlock.md — honest Bayesian update; emits new_posterior +
Δ + recommended_action ∈ {keep, downgrade, upgrade, supersede}.
- write_calibration UPDATEs public.hypotheses + APPENDS a
"## Calibration history" section to the H-NNNN.md case file
(calibration is append-only — each datapoint matters). Posterior
band auto-corrected to match Tetlock thresholds.
- NO_NEW_EVIDENCE sentinel handled; pure 'keep' with |Δ|<0.005 only
touches updated_at + reviewed_by.
- Tone: teal (#26d4cc).
Case-Writer (case_report)
- prompts/case-writer.md — Dr. Watson assembles all artefacts
(E-NNNN, H-NNNN, R-NNNN, W-NNNN, G-NNNN) into a five-act narrative.
ILIKE filter on topic; doc_id optional scope.
- Larger budget cap (≥ $0.50) + longer timeout for prose generation.
- Writes case/reports/<slug>.md with frontmatter (topic + counts);
no DB table for v0.
- New page /c/[slug] renders the report via MarkdownBody + stat chips.
- Tone: gold (#e0c080).
Hardening across the bureau:
- Sentinel parsing now accepts backticked AND prose-trailing forms
(Holmes NO_HYPOTHESES, Dupin NO_CONTRADICTIONS, Schneier
INSUFFICIENT_HYPOTHESIS, Poirot INSUFFICIENT_TESTIMONY, Taleb
NO_OUTLIERS, Tetlock NO_NEW_EVIDENCE, Case-Writer
INSUFFICIENT_ARTEFACTS). Avoids the failure mode where the model
refuses honestly but the runtime treated it as a parse error
(observed live with Poirot+Hoover identifying the DIRECTOR
false-positive disambiguation issue in entity_mentions).
Chat tool extensions (web/lib/chat/tools.ts):
- request_investigation now accepts 7 kinds. Each routes to its
detective with appropriate validation (hypothesis_id regex,
person_id kebab-case, topic non-empty, doc_id for evidence_chain).
- ETA per kind: Holmes/Dupin 60s, Poirot 45s, Schneier/Tetlock 30s,
Taleb 50s, Case-Writer 180s (longer prose), Locard 30×n_chunks.
UI integration:
- chat-bubble inline card paints each detective in its tone color.
- /jobs/[id] page header swaps name/subtitle/tone per detective;
question label adapts ("Topic" / "Hypothesis under attack" /
"Witness under analysis" / "Topic to outlier-scan" / "Hypothesis
under recalibration" / "Case to assemble").
- job-status-poller renders: case-report link card (gold), outlier
cards (yellow), witness cards (purple) — alongside existing
hypothesis, evidence, contradiction cards.
- /api/jobs/[id] hydrates witnesses (JOIN entities for canonical_name)
+ gaps (with scope JSONB).
- /c/[slug] page reads /data/ufo/case/reports/<slug>.md and renders
with MarkdownBody, frontmatter parsed for stat chips.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 01:11:39 +00:00
|
|
|
|
W5.4 (Phase 3B): sitemap + robots + Article schema + magazine reading view
GEO/SEO surface area:
app/robots.ts (new — Next.js dynamic robots)
Explicitly ALLOWS major AI crawlers: GPTBot, OAI-SearchBot,
ChatGPT-User, ClaudeBot, Claude-Web, anthropic-ai, PerplexityBot,
Perplexity-User, Google-Extended, Applebot-Extended, CCBot,
DuckAssistBot, YouBot, Bytespider, Amazonbot. The site exists to
be cited by LLMs answering UAP/UFO questions — we want them in.
/api/admin/, /admin/, /auth/ disallowed for everyone.
app/sitemap.ts (new — Next.js dynamic sitemap)
Lists 9 top-level routes + every /d/<doc> + every /c/<slug> from
the filesystem + up to 500 entity URLs per class
(event, person, uap_object, location, organization),
sorted with summary-enriched entities first. ~3000 URLs total at
current corpus size. lastModified honours summary_generated_at so
crawlers re-index when entities are re-enriched.
app/c/[slug]/page.tsx (rewritten — magazine reading view)
- generateMetadata: per-case title, description (auto-extracted
from the locale-preferred lead paragraph), canonical URL,
hreflang alternate, OpenGraph article type with publishedTime,
Twitter card.
- JSON-LD Article schema embedded at end of page: schema.org
Article + Organization publisher + inLanguage + isAccessibleForFree.
This is what makes the case appear as a citable source in
Google AI Overviews / Perplexity / ChatGPT search.
- Reading view rewritten: display-serif headline (Fraunces), italic
blockquotes with gold accent, prose-typography styling, no more
detective stats line, no more "written by case-writer@detective"
attribution. Locale-aware: PT-BR pulls topic_pt_br + lead in PT,
English mirror.
tailwind.config.ts
+ @tailwindcss/typography plugin
+ font-display family wired to var(--font-display) (Fraunces)
package.json
+ "@tailwindcss/typography" devDependency
Phase 3A note: bulk entity enrichment hit Claude OAuth weekly quota mid-run.
6 events + 3 uap_objects landed bilingual summaries before the quota
exhausted. UI gracefully splits enriched vs bare entities so /sightings
shows the magazine-grade cards (Kenneth Arnold 1947, Roswell, Maury Island,
Joseph Perry 1960 lunar photo, Civil Defense Director 1966, etc.) on top
of a compact table of the rest. Re-run when quota refreshes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 19:09:50 +00:00
|
|
|
const title = locale === "pt-br" ? (fm.topic_pt_br ?? fm.topic ?? slug) : (fm.topic ?? slug);
|
|
|
|
|
const dateLabel = fm.created_at ? new Date(fm.created_at).toLocaleDateString(locale === "pt-br" ? "pt-BR" : "en-US", { day: "numeric", month: "long", year: "numeric" }) : null;
|
|
|
|
|
const canonical = `${SITE_URL}/c/${slug}`;
|
|
|
|
|
const lead = pickLead(body, locale).slice(0, 280);
|
2026-05-24 19:16:20 +00:00
|
|
|
const heroArt = (await hasIllustration(slug)) ? `/api/static/processing/case-art/${slug}.png` : null;
|
W3.8: Investigation Bureau complete — Poirot, Taleb, Tetlock, Case-Writer
Brings the bureau from 4 → 8 detectives. All eight run as Bun + claude-CLI
subprocesses against the same Supabase + investigation_jobs LISTEN/NOTIFY
queue, sharing search.ts hybridSearch and writer-side validators that
gate writes against schema + FK.
New detectives:
Poirot (witness_analysis)
- prompts/poirot.md — credibility / access / bias / corroboration /
verdict; uses entity_mentions JOIN chunks to pull 12 chunks per
person; resolves corroboration_refs chunk_ids defensively (accepts
bare cNNNN even when the model emits pNNN/cNNNN).
- INSERT into public.witnesses with W-NNNN naming.
- Tone: purple (#9b5de5).
Taleb (outlier_scan)
- prompts/taleb.md — "surprise is relative to a model"; at most 3
outliers; each requires explicit dominant_model + why_surprising +
what_it_implies; fan-out into public.gaps with scope.kind="outlier".
- Same unscoped-fallback as Dupin (Pass 1 with doc_id, Pass 2 widens
to corpus if hits < 3).
- Tone: yellow (#ffd23f).
Tetlock (calibrate_hypothesis)
- prompts/tetlock.md — honest Bayesian update; emits new_posterior +
Δ + recommended_action ∈ {keep, downgrade, upgrade, supersede}.
- write_calibration UPDATEs public.hypotheses + APPENDS a
"## Calibration history" section to the H-NNNN.md case file
(calibration is append-only — each datapoint matters). Posterior
band auto-corrected to match Tetlock thresholds.
- NO_NEW_EVIDENCE sentinel handled; pure 'keep' with |Δ|<0.005 only
touches updated_at + reviewed_by.
- Tone: teal (#26d4cc).
Case-Writer (case_report)
- prompts/case-writer.md — Dr. Watson assembles all artefacts
(E-NNNN, H-NNNN, R-NNNN, W-NNNN, G-NNNN) into a five-act narrative.
ILIKE filter on topic; doc_id optional scope.
- Larger budget cap (≥ $0.50) + longer timeout for prose generation.
- Writes case/reports/<slug>.md with frontmatter (topic + counts);
no DB table for v0.
- New page /c/[slug] renders the report via MarkdownBody + stat chips.
- Tone: gold (#e0c080).
Hardening across the bureau:
- Sentinel parsing now accepts backticked AND prose-trailing forms
(Holmes NO_HYPOTHESES, Dupin NO_CONTRADICTIONS, Schneier
INSUFFICIENT_HYPOTHESIS, Poirot INSUFFICIENT_TESTIMONY, Taleb
NO_OUTLIERS, Tetlock NO_NEW_EVIDENCE, Case-Writer
INSUFFICIENT_ARTEFACTS). Avoids the failure mode where the model
refuses honestly but the runtime treated it as a parse error
(observed live with Poirot+Hoover identifying the DIRECTOR
false-positive disambiguation issue in entity_mentions).
Chat tool extensions (web/lib/chat/tools.ts):
- request_investigation now accepts 7 kinds. Each routes to its
detective with appropriate validation (hypothesis_id regex,
person_id kebab-case, topic non-empty, doc_id for evidence_chain).
- ETA per kind: Holmes/Dupin 60s, Poirot 45s, Schneier/Tetlock 30s,
Taleb 50s, Case-Writer 180s (longer prose), Locard 30×n_chunks.
UI integration:
- chat-bubble inline card paints each detective in its tone color.
- /jobs/[id] page header swaps name/subtitle/tone per detective;
question label adapts ("Topic" / "Hypothesis under attack" /
"Witness under analysis" / "Topic to outlier-scan" / "Hypothesis
under recalibration" / "Case to assemble").
- job-status-poller renders: case-report link card (gold), outlier
cards (yellow), witness cards (purple) — alongside existing
hypothesis, evidence, contradiction cards.
- /api/jobs/[id] hydrates witnesses (JOIN entities for canonical_name)
+ gaps (with scope JSONB).
- /c/[slug] page reads /data/ufo/case/reports/<slug>.md and renders
with MarkdownBody, frontmatter parsed for stat chips.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 01:11:39 +00:00
|
|
|
|
|
|
|
|
return (
|
|
|
|
|
<div className="min-h-screen bg-[#0a0e1a] text-[#e7ecf3]">
|
W4.1+W4.2: anti-AI-tics house style + bureau nav (back/home everywhere)
Two complaints in one wave:
(W4.1) User: "Não pode ter vícios de IA como uso excessivo de '-' que a IA
coloca geralmente no lugar de vírgulas por exemplo. Isso deve fazer parte
do prompt geral."
- New prompts/_house-style.md banning the 9 most common AI prose tells
in both EN and PT-BR:
1. Em dashes as comma replacements (—)
2. Rule-of-three lists ("concrete, rigorous, and grounded")
3. Conjunctive openers ("Moreover", "Notably", "Ademais")
4. Superficial -ing analyses ("marking a shift", "destacando")
5. Inflated symbolism + AI vocab (tapestry, navigate, delve,
underscore, robust, multifaceted, marco histórico, ...)
6. Negative parallelisms ("Not just X but Y")
7. Vague attribution ("Some scholars say...")
8. Summary closers ("In summary...", "Em suma...")
9. Hedging fluff ("It's important to note...")
Verbatim chunk quotes are explicitly exempt; preserve as-is.
- claude.ts callClaude() lazily loads _house-style.md once per process
and PREPENDS it to every detective's system prompt:
composedSystem = houseStyle + "---" + detective.systemPrompt
This means all 7 detectives + future ones get the rules without any
per-prompt change.
(W4.2) User: "Quando entra em uma página da investigação não tem como
voltar! UX terrível!"
- New <BureauNav> sticky topbar with explicit "← home" + "🔎 bureau"
buttons + clickable breadcrumb trail. Always visible at the top of
every bureau page so the user can escape in one click.
- Wired into /bureau, /h/[hypothesisId], /c/[slug], /jobs/[id]. Each
page passes its sensible parent crumb (/bureau#hypotheses,
/bureau#reports, /bureau#jobs).
- Replaces the previous plain-text "disclosure.top / hypothesis /
H-0004" line which had no visual affordance.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 16:27:58 +00:00
|
|
|
<BureauNav crumbs={[
|
W5.4 (Phase 3B): sitemap + robots + Article schema + magazine reading view
GEO/SEO surface area:
app/robots.ts (new — Next.js dynamic robots)
Explicitly ALLOWS major AI crawlers: GPTBot, OAI-SearchBot,
ChatGPT-User, ClaudeBot, Claude-Web, anthropic-ai, PerplexityBot,
Perplexity-User, Google-Extended, Applebot-Extended, CCBot,
DuckAssistBot, YouBot, Bytespider, Amazonbot. The site exists to
be cited by LLMs answering UAP/UFO questions — we want them in.
/api/admin/, /admin/, /auth/ disallowed for everyone.
app/sitemap.ts (new — Next.js dynamic sitemap)
Lists 9 top-level routes + every /d/<doc> + every /c/<slug> from
the filesystem + up to 500 entity URLs per class
(event, person, uap_object, location, organization),
sorted with summary-enriched entities first. ~3000 URLs total at
current corpus size. lastModified honours summary_generated_at so
crawlers re-index when entities are re-enriched.
app/c/[slug]/page.tsx (rewritten — magazine reading view)
- generateMetadata: per-case title, description (auto-extracted
from the locale-preferred lead paragraph), canonical URL,
hreflang alternate, OpenGraph article type with publishedTime,
Twitter card.
- JSON-LD Article schema embedded at end of page: schema.org
Article + Organization publisher + inLanguage + isAccessibleForFree.
This is what makes the case appear as a citable source in
Google AI Overviews / Perplexity / ChatGPT search.
- Reading view rewritten: display-serif headline (Fraunces), italic
blockquotes with gold accent, prose-typography styling, no more
detective stats line, no more "written by case-writer@detective"
attribution. Locale-aware: PT-BR pulls topic_pt_br + lead in PT,
English mirror.
tailwind.config.ts
+ @tailwindcss/typography plugin
+ font-display family wired to var(--font-display) (Fraunces)
package.json
+ "@tailwindcss/typography" devDependency
Phase 3A note: bulk entity enrichment hit Claude OAuth weekly quota mid-run.
6 events + 3 uap_objects landed bilingual summaries before the quota
exhausted. UI gracefully splits enriched vs bare entities so /sightings
shows the magazine-grade cards (Kenneth Arnold 1947, Roswell, Maury Island,
Joseph Perry 1960 lunar photo, Civil Defense Director 1966, etc.) on top
of a compact table of the rest. Re-run when quota refreshes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 19:09:50 +00:00
|
|
|
{ label: locale === "en" ? "case files" : "casos", href: "/bureau" },
|
|
|
|
|
{ label: title.length > 32 ? title.slice(0, 32) + "…" : title },
|
W4.1+W4.2: anti-AI-tics house style + bureau nav (back/home everywhere)
Two complaints in one wave:
(W4.1) User: "Não pode ter vícios de IA como uso excessivo de '-' que a IA
coloca geralmente no lugar de vírgulas por exemplo. Isso deve fazer parte
do prompt geral."
- New prompts/_house-style.md banning the 9 most common AI prose tells
in both EN and PT-BR:
1. Em dashes as comma replacements (—)
2. Rule-of-three lists ("concrete, rigorous, and grounded")
3. Conjunctive openers ("Moreover", "Notably", "Ademais")
4. Superficial -ing analyses ("marking a shift", "destacando")
5. Inflated symbolism + AI vocab (tapestry, navigate, delve,
underscore, robust, multifaceted, marco histórico, ...)
6. Negative parallelisms ("Not just X but Y")
7. Vague attribution ("Some scholars say...")
8. Summary closers ("In summary...", "Em suma...")
9. Hedging fluff ("It's important to note...")
Verbatim chunk quotes are explicitly exempt; preserve as-is.
- claude.ts callClaude() lazily loads _house-style.md once per process
and PREPENDS it to every detective's system prompt:
composedSystem = houseStyle + "---" + detective.systemPrompt
This means all 7 detectives + future ones get the rules without any
per-prompt change.
(W4.2) User: "Quando entra em uma página da investigação não tem como
voltar! UX terrível!"
- New <BureauNav> sticky topbar with explicit "← home" + "🔎 bureau"
buttons + clickable breadcrumb trail. Always visible at the top of
every bureau page so the user can escape in one click.
- Wired into /bureau, /h/[hypothesisId], /c/[slug], /jobs/[id]. Each
page passes its sensible parent crumb (/bureau#hypotheses,
/bureau#reports, /bureau#jobs).
- Replaces the previous plain-text "disclosure.top / hypothesis /
H-0004" line which had no visual affordance.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 16:27:58 +00:00
|
|
|
]} />
|
W3.8: Investigation Bureau complete — Poirot, Taleb, Tetlock, Case-Writer
Brings the bureau from 4 → 8 detectives. All eight run as Bun + claude-CLI
subprocesses against the same Supabase + investigation_jobs LISTEN/NOTIFY
queue, sharing search.ts hybridSearch and writer-side validators that
gate writes against schema + FK.
New detectives:
Poirot (witness_analysis)
- prompts/poirot.md — credibility / access / bias / corroboration /
verdict; uses entity_mentions JOIN chunks to pull 12 chunks per
person; resolves corroboration_refs chunk_ids defensively (accepts
bare cNNNN even when the model emits pNNN/cNNNN).
- INSERT into public.witnesses with W-NNNN naming.
- Tone: purple (#9b5de5).
Taleb (outlier_scan)
- prompts/taleb.md — "surprise is relative to a model"; at most 3
outliers; each requires explicit dominant_model + why_surprising +
what_it_implies; fan-out into public.gaps with scope.kind="outlier".
- Same unscoped-fallback as Dupin (Pass 1 with doc_id, Pass 2 widens
to corpus if hits < 3).
- Tone: yellow (#ffd23f).
Tetlock (calibrate_hypothesis)
- prompts/tetlock.md — honest Bayesian update; emits new_posterior +
Δ + recommended_action ∈ {keep, downgrade, upgrade, supersede}.
- write_calibration UPDATEs public.hypotheses + APPENDS a
"## Calibration history" section to the H-NNNN.md case file
(calibration is append-only — each datapoint matters). Posterior
band auto-corrected to match Tetlock thresholds.
- NO_NEW_EVIDENCE sentinel handled; pure 'keep' with |Δ|<0.005 only
touches updated_at + reviewed_by.
- Tone: teal (#26d4cc).
Case-Writer (case_report)
- prompts/case-writer.md — Dr. Watson assembles all artefacts
(E-NNNN, H-NNNN, R-NNNN, W-NNNN, G-NNNN) into a five-act narrative.
ILIKE filter on topic; doc_id optional scope.
- Larger budget cap (≥ $0.50) + longer timeout for prose generation.
- Writes case/reports/<slug>.md with frontmatter (topic + counts);
no DB table for v0.
- New page /c/[slug] renders the report via MarkdownBody + stat chips.
- Tone: gold (#e0c080).
Hardening across the bureau:
- Sentinel parsing now accepts backticked AND prose-trailing forms
(Holmes NO_HYPOTHESES, Dupin NO_CONTRADICTIONS, Schneier
INSUFFICIENT_HYPOTHESIS, Poirot INSUFFICIENT_TESTIMONY, Taleb
NO_OUTLIERS, Tetlock NO_NEW_EVIDENCE, Case-Writer
INSUFFICIENT_ARTEFACTS). Avoids the failure mode where the model
refuses honestly but the runtime treated it as a parse error
(observed live with Poirot+Hoover identifying the DIRECTOR
false-positive disambiguation issue in entity_mentions).
Chat tool extensions (web/lib/chat/tools.ts):
- request_investigation now accepts 7 kinds. Each routes to its
detective with appropriate validation (hypothesis_id regex,
person_id kebab-case, topic non-empty, doc_id for evidence_chain).
- ETA per kind: Holmes/Dupin 60s, Poirot 45s, Schneier/Tetlock 30s,
Taleb 50s, Case-Writer 180s (longer prose), Locard 30×n_chunks.
UI integration:
- chat-bubble inline card paints each detective in its tone color.
- /jobs/[id] page header swaps name/subtitle/tone per detective;
question label adapts ("Topic" / "Hypothesis under attack" /
"Witness under analysis" / "Topic to outlier-scan" / "Hypothesis
under recalibration" / "Case to assemble").
- job-status-poller renders: case-report link card (gold), outlier
cards (yellow), witness cards (purple) — alongside existing
hypothesis, evidence, contradiction cards.
- /api/jobs/[id] hydrates witnesses (JOIN entities for canonical_name)
+ gaps (with scope JSONB).
- /c/[slug] page reads /data/ufo/case/reports/<slug>.md and renders
with MarkdownBody, frontmatter parsed for stat chips.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 01:11:39 +00:00
|
|
|
<AuthBar />
|
|
|
|
|
|
2026-05-24 19:16:20 +00:00
|
|
|
{/* Full-bleed editorial hero — only when a painterly illustration
|
|
|
|
|
exists for this case. Other cases get the plain title header. */}
|
|
|
|
|
{heroArt && (
|
|
|
|
|
<div className="relative w-full overflow-hidden border-b border-[rgba(224,192,128,0.15)]">
|
|
|
|
|
<div className="relative aspect-[16/9] md:aspect-[21/9] max-h-[640px]">
|
|
|
|
|
{/* eslint-disable-next-line @next/next/no-img-element */}
|
|
|
|
|
<img
|
|
|
|
|
src={heroArt}
|
|
|
|
|
alt={title}
|
|
|
|
|
className="absolute inset-0 w-full h-full object-cover"
|
|
|
|
|
/>
|
|
|
|
|
<div className="absolute inset-0 bg-gradient-to-t from-[#0a0e1a] via-[#0a0e1a]/30 to-transparent" />
|
|
|
|
|
<div className="absolute inset-0 bg-gradient-to-r from-[#0a0e1a]/70 via-transparent to-transparent" />
|
|
|
|
|
<div className="absolute bottom-0 left-0 right-0 px-4 md:px-8 pb-8 md:pb-12">
|
|
|
|
|
<div className="mx-auto max-w-3xl">
|
|
|
|
|
<div className="text-[10px] font-mono uppercase tracking-[0.18em] text-[#e0c080] mb-3">
|
|
|
|
|
{locale === "en" ? "Declassified case file" : "Arquivo desclassificado"}
|
|
|
|
|
{dateLabel && <> · {dateLabel}</>}
|
|
|
|
|
</div>
|
|
|
|
|
<h1 className="font-display text-4xl md:text-6xl font-semibold leading-[1.05] tracking-tight text-white drop-shadow-lg">
|
|
|
|
|
{title}
|
|
|
|
|
</h1>
|
|
|
|
|
</div>
|
|
|
|
|
</div>
|
|
|
|
|
<div className="absolute top-3 right-3 text-[9px] font-mono uppercase tracking-wider text-[#9aa6b8]/80 bg-[#0a0e1a]/70 px-2 py-1 rounded">
|
|
|
|
|
{locale === "en" ? "Editorial illustration" : "Ilustração editorial"}
|
|
|
|
|
</div>
|
W3.8: Investigation Bureau complete — Poirot, Taleb, Tetlock, Case-Writer
Brings the bureau from 4 → 8 detectives. All eight run as Bun + claude-CLI
subprocesses against the same Supabase + investigation_jobs LISTEN/NOTIFY
queue, sharing search.ts hybridSearch and writer-side validators that
gate writes against schema + FK.
New detectives:
Poirot (witness_analysis)
- prompts/poirot.md — credibility / access / bias / corroboration /
verdict; uses entity_mentions JOIN chunks to pull 12 chunks per
person; resolves corroboration_refs chunk_ids defensively (accepts
bare cNNNN even when the model emits pNNN/cNNNN).
- INSERT into public.witnesses with W-NNNN naming.
- Tone: purple (#9b5de5).
Taleb (outlier_scan)
- prompts/taleb.md — "surprise is relative to a model"; at most 3
outliers; each requires explicit dominant_model + why_surprising +
what_it_implies; fan-out into public.gaps with scope.kind="outlier".
- Same unscoped-fallback as Dupin (Pass 1 with doc_id, Pass 2 widens
to corpus if hits < 3).
- Tone: yellow (#ffd23f).
Tetlock (calibrate_hypothesis)
- prompts/tetlock.md — honest Bayesian update; emits new_posterior +
Δ + recommended_action ∈ {keep, downgrade, upgrade, supersede}.
- write_calibration UPDATEs public.hypotheses + APPENDS a
"## Calibration history" section to the H-NNNN.md case file
(calibration is append-only — each datapoint matters). Posterior
band auto-corrected to match Tetlock thresholds.
- NO_NEW_EVIDENCE sentinel handled; pure 'keep' with |Δ|<0.005 only
touches updated_at + reviewed_by.
- Tone: teal (#26d4cc).
Case-Writer (case_report)
- prompts/case-writer.md — Dr. Watson assembles all artefacts
(E-NNNN, H-NNNN, R-NNNN, W-NNNN, G-NNNN) into a five-act narrative.
ILIKE filter on topic; doc_id optional scope.
- Larger budget cap (≥ $0.50) + longer timeout for prose generation.
- Writes case/reports/<slug>.md with frontmatter (topic + counts);
no DB table for v0.
- New page /c/[slug] renders the report via MarkdownBody + stat chips.
- Tone: gold (#e0c080).
Hardening across the bureau:
- Sentinel parsing now accepts backticked AND prose-trailing forms
(Holmes NO_HYPOTHESES, Dupin NO_CONTRADICTIONS, Schneier
INSUFFICIENT_HYPOTHESIS, Poirot INSUFFICIENT_TESTIMONY, Taleb
NO_OUTLIERS, Tetlock NO_NEW_EVIDENCE, Case-Writer
INSUFFICIENT_ARTEFACTS). Avoids the failure mode where the model
refuses honestly but the runtime treated it as a parse error
(observed live with Poirot+Hoover identifying the DIRECTOR
false-positive disambiguation issue in entity_mentions).
Chat tool extensions (web/lib/chat/tools.ts):
- request_investigation now accepts 7 kinds. Each routes to its
detective with appropriate validation (hypothesis_id regex,
person_id kebab-case, topic non-empty, doc_id for evidence_chain).
- ETA per kind: Holmes/Dupin 60s, Poirot 45s, Schneier/Tetlock 30s,
Taleb 50s, Case-Writer 180s (longer prose), Locard 30×n_chunks.
UI integration:
- chat-bubble inline card paints each detective in its tone color.
- /jobs/[id] page header swaps name/subtitle/tone per detective;
question label adapts ("Topic" / "Hypothesis under attack" /
"Witness under analysis" / "Topic to outlier-scan" / "Hypothesis
under recalibration" / "Case to assemble").
- job-status-poller renders: case-report link card (gold), outlier
cards (yellow), witness cards (purple) — alongside existing
hypothesis, evidence, contradiction cards.
- /api/jobs/[id] hydrates witnesses (JOIN entities for canonical_name)
+ gaps (with scope JSONB).
- /c/[slug] page reads /data/ufo/case/reports/<slug>.md and renders
with MarkdownBody, frontmatter parsed for stat chips.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 01:11:39 +00:00
|
|
|
</div>
|
2026-05-24 19:16:20 +00:00
|
|
|
</div>
|
|
|
|
|
)}
|
|
|
|
|
|
|
|
|
|
<article className="mx-auto max-w-3xl px-4 py-10 md:py-14">
|
|
|
|
|
{!heroArt && (
|
|
|
|
|
<header className="mb-10 md:mb-14">
|
|
|
|
|
<div className="text-[10px] font-mono uppercase tracking-[0.18em] text-[#e0c080] mb-4">
|
|
|
|
|
{locale === "en" ? "Declassified case file" : "Arquivo desclassificado"}
|
|
|
|
|
{dateLabel && <> · {dateLabel}</>}
|
|
|
|
|
</div>
|
|
|
|
|
<h1 className="font-display text-4xl md:text-6xl font-semibold leading-[1.05] tracking-tight text-[#e7ecf3] mb-6">
|
|
|
|
|
{title}
|
|
|
|
|
</h1>
|
|
|
|
|
</header>
|
|
|
|
|
)}
|
|
|
|
|
{lead && (
|
|
|
|
|
<p className={`text-lg md:text-xl text-[#cbd2dd] leading-relaxed font-light max-w-2xl ${heroArt ? "mb-10" : "mb-10 md:mb-14"}`}>
|
|
|
|
|
{lead}
|
|
|
|
|
</p>
|
|
|
|
|
)}
|
W3.8: Investigation Bureau complete — Poirot, Taleb, Tetlock, Case-Writer
Brings the bureau from 4 → 8 detectives. All eight run as Bun + claude-CLI
subprocesses against the same Supabase + investigation_jobs LISTEN/NOTIFY
queue, sharing search.ts hybridSearch and writer-side validators that
gate writes against schema + FK.
New detectives:
Poirot (witness_analysis)
- prompts/poirot.md — credibility / access / bias / corroboration /
verdict; uses entity_mentions JOIN chunks to pull 12 chunks per
person; resolves corroboration_refs chunk_ids defensively (accepts
bare cNNNN even when the model emits pNNN/cNNNN).
- INSERT into public.witnesses with W-NNNN naming.
- Tone: purple (#9b5de5).
Taleb (outlier_scan)
- prompts/taleb.md — "surprise is relative to a model"; at most 3
outliers; each requires explicit dominant_model + why_surprising +
what_it_implies; fan-out into public.gaps with scope.kind="outlier".
- Same unscoped-fallback as Dupin (Pass 1 with doc_id, Pass 2 widens
to corpus if hits < 3).
- Tone: yellow (#ffd23f).
Tetlock (calibrate_hypothesis)
- prompts/tetlock.md — honest Bayesian update; emits new_posterior +
Δ + recommended_action ∈ {keep, downgrade, upgrade, supersede}.
- write_calibration UPDATEs public.hypotheses + APPENDS a
"## Calibration history" section to the H-NNNN.md case file
(calibration is append-only — each datapoint matters). Posterior
band auto-corrected to match Tetlock thresholds.
- NO_NEW_EVIDENCE sentinel handled; pure 'keep' with |Δ|<0.005 only
touches updated_at + reviewed_by.
- Tone: teal (#26d4cc).
Case-Writer (case_report)
- prompts/case-writer.md — Dr. Watson assembles all artefacts
(E-NNNN, H-NNNN, R-NNNN, W-NNNN, G-NNNN) into a five-act narrative.
ILIKE filter on topic; doc_id optional scope.
- Larger budget cap (≥ $0.50) + longer timeout for prose generation.
- Writes case/reports/<slug>.md with frontmatter (topic + counts);
no DB table for v0.
- New page /c/[slug] renders the report via MarkdownBody + stat chips.
- Tone: gold (#e0c080).
Hardening across the bureau:
- Sentinel parsing now accepts backticked AND prose-trailing forms
(Holmes NO_HYPOTHESES, Dupin NO_CONTRADICTIONS, Schneier
INSUFFICIENT_HYPOTHESIS, Poirot INSUFFICIENT_TESTIMONY, Taleb
NO_OUTLIERS, Tetlock NO_NEW_EVIDENCE, Case-Writer
INSUFFICIENT_ARTEFACTS). Avoids the failure mode where the model
refuses honestly but the runtime treated it as a parse error
(observed live with Poirot+Hoover identifying the DIRECTOR
false-positive disambiguation issue in entity_mentions).
Chat tool extensions (web/lib/chat/tools.ts):
- request_investigation now accepts 7 kinds. Each routes to its
detective with appropriate validation (hypothesis_id regex,
person_id kebab-case, topic non-empty, doc_id for evidence_chain).
- ETA per kind: Holmes/Dupin 60s, Poirot 45s, Schneier/Tetlock 30s,
Taleb 50s, Case-Writer 180s (longer prose), Locard 30×n_chunks.
UI integration:
- chat-bubble inline card paints each detective in its tone color.
- /jobs/[id] page header swaps name/subtitle/tone per detective;
question label adapts ("Topic" / "Hypothesis under attack" /
"Witness under analysis" / "Topic to outlier-scan" / "Hypothesis
under recalibration" / "Case to assemble").
- job-status-poller renders: case-report link card (gold), outlier
cards (yellow), witness cards (purple) — alongside existing
hypothesis, evidence, contradiction cards.
- /api/jobs/[id] hydrates witnesses (JOIN entities for canonical_name)
+ gaps (with scope JSONB).
- /c/[slug] page reads /data/ufo/case/reports/<slug>.md and renders
with MarkdownBody, frontmatter parsed for stat chips.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 01:11:39 +00:00
|
|
|
|
W5.4 (Phase 3B): sitemap + robots + Article schema + magazine reading view
GEO/SEO surface area:
app/robots.ts (new — Next.js dynamic robots)
Explicitly ALLOWS major AI crawlers: GPTBot, OAI-SearchBot,
ChatGPT-User, ClaudeBot, Claude-Web, anthropic-ai, PerplexityBot,
Perplexity-User, Google-Extended, Applebot-Extended, CCBot,
DuckAssistBot, YouBot, Bytespider, Amazonbot. The site exists to
be cited by LLMs answering UAP/UFO questions — we want them in.
/api/admin/, /admin/, /auth/ disallowed for everyone.
app/sitemap.ts (new — Next.js dynamic sitemap)
Lists 9 top-level routes + every /d/<doc> + every /c/<slug> from
the filesystem + up to 500 entity URLs per class
(event, person, uap_object, location, organization),
sorted with summary-enriched entities first. ~3000 URLs total at
current corpus size. lastModified honours summary_generated_at so
crawlers re-index when entities are re-enriched.
app/c/[slug]/page.tsx (rewritten — magazine reading view)
- generateMetadata: per-case title, description (auto-extracted
from the locale-preferred lead paragraph), canonical URL,
hreflang alternate, OpenGraph article type with publishedTime,
Twitter card.
- JSON-LD Article schema embedded at end of page: schema.org
Article + Organization publisher + inLanguage + isAccessibleForFree.
This is what makes the case appear as a citable source in
Google AI Overviews / Perplexity / ChatGPT search.
- Reading view rewritten: display-serif headline (Fraunces), italic
blockquotes with gold accent, prose-typography styling, no more
detective stats line, no more "written by case-writer@detective"
attribution. Locale-aware: PT-BR pulls topic_pt_br + lead in PT,
English mirror.
tailwind.config.ts
+ @tailwindcss/typography plugin
+ font-display family wired to var(--font-display) (Fraunces)
package.json
+ "@tailwindcss/typography" devDependency
Phase 3A note: bulk entity enrichment hit Claude OAuth weekly quota mid-run.
6 events + 3 uap_objects landed bilingual summaries before the quota
exhausted. UI gracefully splits enriched vs bare entities so /sightings
shows the magazine-grade cards (Kenneth Arnold 1947, Roswell, Maury Island,
Joseph Perry 1960 lunar photo, Civil Defense Director 1966, etc.) on top
of a compact table of the rest. Re-run when quota refreshes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 19:09:50 +00:00
|
|
|
<div className="prose prose-invert prose-lg max-w-none
|
|
|
|
|
prose-headings:font-display prose-headings:font-semibold
|
|
|
|
|
prose-h2:text-2xl prose-h2:md:text-3xl prose-h2:mt-12 prose-h2:mb-4
|
|
|
|
|
prose-h2:border-b prose-h2:border-[rgba(224,192,128,0.15)] prose-h2:pb-3
|
|
|
|
|
prose-p:text-[16px] prose-p:leading-relaxed prose-p:text-[#cbd2dd]
|
|
|
|
|
prose-blockquote:border-l-[#e0c080] prose-blockquote:bg-[rgba(224,192,128,0.05)]
|
|
|
|
|
prose-blockquote:not-italic prose-blockquote:font-display prose-blockquote:text-[#e7ecf3]
|
|
|
|
|
prose-blockquote:py-2 prose-blockquote:px-4 prose-blockquote:rounded-r
|
|
|
|
|
prose-a:text-[#7fdbff] prose-a:no-underline hover:prose-a:underline
|
|
|
|
|
prose-strong:text-[#e7ecf3]">
|
W3.8: Investigation Bureau complete — Poirot, Taleb, Tetlock, Case-Writer
Brings the bureau from 4 → 8 detectives. All eight run as Bun + claude-CLI
subprocesses against the same Supabase + investigation_jobs LISTEN/NOTIFY
queue, sharing search.ts hybridSearch and writer-side validators that
gate writes against schema + FK.
New detectives:
Poirot (witness_analysis)
- prompts/poirot.md — credibility / access / bias / corroboration /
verdict; uses entity_mentions JOIN chunks to pull 12 chunks per
person; resolves corroboration_refs chunk_ids defensively (accepts
bare cNNNN even when the model emits pNNN/cNNNN).
- INSERT into public.witnesses with W-NNNN naming.
- Tone: purple (#9b5de5).
Taleb (outlier_scan)
- prompts/taleb.md — "surprise is relative to a model"; at most 3
outliers; each requires explicit dominant_model + why_surprising +
what_it_implies; fan-out into public.gaps with scope.kind="outlier".
- Same unscoped-fallback as Dupin (Pass 1 with doc_id, Pass 2 widens
to corpus if hits < 3).
- Tone: yellow (#ffd23f).
Tetlock (calibrate_hypothesis)
- prompts/tetlock.md — honest Bayesian update; emits new_posterior +
Δ + recommended_action ∈ {keep, downgrade, upgrade, supersede}.
- write_calibration UPDATEs public.hypotheses + APPENDS a
"## Calibration history" section to the H-NNNN.md case file
(calibration is append-only — each datapoint matters). Posterior
band auto-corrected to match Tetlock thresholds.
- NO_NEW_EVIDENCE sentinel handled; pure 'keep' with |Δ|<0.005 only
touches updated_at + reviewed_by.
- Tone: teal (#26d4cc).
Case-Writer (case_report)
- prompts/case-writer.md — Dr. Watson assembles all artefacts
(E-NNNN, H-NNNN, R-NNNN, W-NNNN, G-NNNN) into a five-act narrative.
ILIKE filter on topic; doc_id optional scope.
- Larger budget cap (≥ $0.50) + longer timeout for prose generation.
- Writes case/reports/<slug>.md with frontmatter (topic + counts);
no DB table for v0.
- New page /c/[slug] renders the report via MarkdownBody + stat chips.
- Tone: gold (#e0c080).
Hardening across the bureau:
- Sentinel parsing now accepts backticked AND prose-trailing forms
(Holmes NO_HYPOTHESES, Dupin NO_CONTRADICTIONS, Schneier
INSUFFICIENT_HYPOTHESIS, Poirot INSUFFICIENT_TESTIMONY, Taleb
NO_OUTLIERS, Tetlock NO_NEW_EVIDENCE, Case-Writer
INSUFFICIENT_ARTEFACTS). Avoids the failure mode where the model
refuses honestly but the runtime treated it as a parse error
(observed live with Poirot+Hoover identifying the DIRECTOR
false-positive disambiguation issue in entity_mentions).
Chat tool extensions (web/lib/chat/tools.ts):
- request_investigation now accepts 7 kinds. Each routes to its
detective with appropriate validation (hypothesis_id regex,
person_id kebab-case, topic non-empty, doc_id for evidence_chain).
- ETA per kind: Holmes/Dupin 60s, Poirot 45s, Schneier/Tetlock 30s,
Taleb 50s, Case-Writer 180s (longer prose), Locard 30×n_chunks.
UI integration:
- chat-bubble inline card paints each detective in its tone color.
- /jobs/[id] page header swaps name/subtitle/tone per detective;
question label adapts ("Topic" / "Hypothesis under attack" /
"Witness under analysis" / "Topic to outlier-scan" / "Hypothesis
under recalibration" / "Case to assemble").
- job-status-poller renders: case-report link card (gold), outlier
cards (yellow), witness cards (purple) — alongside existing
hypothesis, evidence, contradiction cards.
- /api/jobs/[id] hydrates witnesses (JOIN entities for canonical_name)
+ gaps (with scope JSONB).
- /c/[slug] page reads /data/ufo/case/reports/<slug>.md and renders
with MarkdownBody, frontmatter parsed for stat chips.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 01:11:39 +00:00
|
|
|
<MarkdownBody>{body}</MarkdownBody>
|
W5.4 (Phase 3B): sitemap + robots + Article schema + magazine reading view
GEO/SEO surface area:
app/robots.ts (new — Next.js dynamic robots)
Explicitly ALLOWS major AI crawlers: GPTBot, OAI-SearchBot,
ChatGPT-User, ClaudeBot, Claude-Web, anthropic-ai, PerplexityBot,
Perplexity-User, Google-Extended, Applebot-Extended, CCBot,
DuckAssistBot, YouBot, Bytespider, Amazonbot. The site exists to
be cited by LLMs answering UAP/UFO questions — we want them in.
/api/admin/, /admin/, /auth/ disallowed for everyone.
app/sitemap.ts (new — Next.js dynamic sitemap)
Lists 9 top-level routes + every /d/<doc> + every /c/<slug> from
the filesystem + up to 500 entity URLs per class
(event, person, uap_object, location, organization),
sorted with summary-enriched entities first. ~3000 URLs total at
current corpus size. lastModified honours summary_generated_at so
crawlers re-index when entities are re-enriched.
app/c/[slug]/page.tsx (rewritten — magazine reading view)
- generateMetadata: per-case title, description (auto-extracted
from the locale-preferred lead paragraph), canonical URL,
hreflang alternate, OpenGraph article type with publishedTime,
Twitter card.
- JSON-LD Article schema embedded at end of page: schema.org
Article + Organization publisher + inLanguage + isAccessibleForFree.
This is what makes the case appear as a citable source in
Google AI Overviews / Perplexity / ChatGPT search.
- Reading view rewritten: display-serif headline (Fraunces), italic
blockquotes with gold accent, prose-typography styling, no more
detective stats line, no more "written by case-writer@detective"
attribution. Locale-aware: PT-BR pulls topic_pt_br + lead in PT,
English mirror.
tailwind.config.ts
+ @tailwindcss/typography plugin
+ font-display family wired to var(--font-display) (Fraunces)
package.json
+ "@tailwindcss/typography" devDependency
Phase 3A note: bulk entity enrichment hit Claude OAuth weekly quota mid-run.
6 events + 3 uap_objects landed bilingual summaries before the quota
exhausted. UI gracefully splits enriched vs bare entities so /sightings
shows the magazine-grade cards (Kenneth Arnold 1947, Roswell, Maury Island,
Joseph Perry 1960 lunar photo, Civil Defense Director 1966, etc.) on top
of a compact table of the rest. Re-run when quota refreshes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 19:09:50 +00:00
|
|
|
</div>
|
|
|
|
|
</article>
|
|
|
|
|
|
|
|
|
|
{/* JSON-LD Article — helps Google + AI crawlers parse the case as
|
|
|
|
|
a citation-bearing piece of journalism */}
|
|
|
|
|
<script
|
|
|
|
|
type="application/ld+json"
|
|
|
|
|
dangerouslySetInnerHTML={{ __html: JSON.stringify({
|
|
|
|
|
"@context": "https://schema.org",
|
|
|
|
|
"@type": "Article",
|
|
|
|
|
headline: title,
|
|
|
|
|
description: lead,
|
|
|
|
|
url: canonical,
|
|
|
|
|
datePublished: fm.created_at,
|
|
|
|
|
inLanguage: locale === "pt-br" ? "pt-BR" : "en-US",
|
|
|
|
|
isAccessibleForFree: true,
|
|
|
|
|
publisher: {
|
|
|
|
|
"@type": "Organization",
|
|
|
|
|
name: "The Disclosure Bureau",
|
|
|
|
|
url: SITE_URL,
|
|
|
|
|
},
|
|
|
|
|
about: { "@type": "Thing", name: "UAP/UFO declassified record" },
|
|
|
|
|
}) }}
|
|
|
|
|
/>
|
W3.8: Investigation Bureau complete — Poirot, Taleb, Tetlock, Case-Writer
Brings the bureau from 4 → 8 detectives. All eight run as Bun + claude-CLI
subprocesses against the same Supabase + investigation_jobs LISTEN/NOTIFY
queue, sharing search.ts hybridSearch and writer-side validators that
gate writes against schema + FK.
New detectives:
Poirot (witness_analysis)
- prompts/poirot.md — credibility / access / bias / corroboration /
verdict; uses entity_mentions JOIN chunks to pull 12 chunks per
person; resolves corroboration_refs chunk_ids defensively (accepts
bare cNNNN even when the model emits pNNN/cNNNN).
- INSERT into public.witnesses with W-NNNN naming.
- Tone: purple (#9b5de5).
Taleb (outlier_scan)
- prompts/taleb.md — "surprise is relative to a model"; at most 3
outliers; each requires explicit dominant_model + why_surprising +
what_it_implies; fan-out into public.gaps with scope.kind="outlier".
- Same unscoped-fallback as Dupin (Pass 1 with doc_id, Pass 2 widens
to corpus if hits < 3).
- Tone: yellow (#ffd23f).
Tetlock (calibrate_hypothesis)
- prompts/tetlock.md — honest Bayesian update; emits new_posterior +
Δ + recommended_action ∈ {keep, downgrade, upgrade, supersede}.
- write_calibration UPDATEs public.hypotheses + APPENDS a
"## Calibration history" section to the H-NNNN.md case file
(calibration is append-only — each datapoint matters). Posterior
band auto-corrected to match Tetlock thresholds.
- NO_NEW_EVIDENCE sentinel handled; pure 'keep' with |Δ|<0.005 only
touches updated_at + reviewed_by.
- Tone: teal (#26d4cc).
Case-Writer (case_report)
- prompts/case-writer.md — Dr. Watson assembles all artefacts
(E-NNNN, H-NNNN, R-NNNN, W-NNNN, G-NNNN) into a five-act narrative.
ILIKE filter on topic; doc_id optional scope.
- Larger budget cap (≥ $0.50) + longer timeout for prose generation.
- Writes case/reports/<slug>.md with frontmatter (topic + counts);
no DB table for v0.
- New page /c/[slug] renders the report via MarkdownBody + stat chips.
- Tone: gold (#e0c080).
Hardening across the bureau:
- Sentinel parsing now accepts backticked AND prose-trailing forms
(Holmes NO_HYPOTHESES, Dupin NO_CONTRADICTIONS, Schneier
INSUFFICIENT_HYPOTHESIS, Poirot INSUFFICIENT_TESTIMONY, Taleb
NO_OUTLIERS, Tetlock NO_NEW_EVIDENCE, Case-Writer
INSUFFICIENT_ARTEFACTS). Avoids the failure mode where the model
refuses honestly but the runtime treated it as a parse error
(observed live with Poirot+Hoover identifying the DIRECTOR
false-positive disambiguation issue in entity_mentions).
Chat tool extensions (web/lib/chat/tools.ts):
- request_investigation now accepts 7 kinds. Each routes to its
detective with appropriate validation (hypothesis_id regex,
person_id kebab-case, topic non-empty, doc_id for evidence_chain).
- ETA per kind: Holmes/Dupin 60s, Poirot 45s, Schneier/Tetlock 30s,
Taleb 50s, Case-Writer 180s (longer prose), Locard 30×n_chunks.
UI integration:
- chat-bubble inline card paints each detective in its tone color.
- /jobs/[id] page header swaps name/subtitle/tone per detective;
question label adapts ("Topic" / "Hypothesis under attack" /
"Witness under analysis" / "Topic to outlier-scan" / "Hypothesis
under recalibration" / "Case to assemble").
- job-status-poller renders: case-report link card (gold), outlier
cards (yellow), witness cards (purple) — alongside existing
hypothesis, evidence, contradiction cards.
- /api/jobs/[id] hydrates witnesses (JOIN entities for canonical_name)
+ gaps (with scope JSONB).
- /c/[slug] page reads /data/ufo/case/reports/<slug>.md and renders
with MarkdownBody, frontmatter parsed for stat chips.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-24 01:11:39 +00:00
|
|
|
</div>
|
|
|
|
|
);
|
|
|
|
|
}
|