morningstar-court-reporter

name

morningstar-court-reporter

description

Integrates and updates all MORNINGSTAR courtroom documentation sources—transcripts, precedents, metrics, dashboard, manifest, state, CHANGELOG. Authenticates and renames transcripts per core/case-format.md. Verifies winddown checklist completion. Use when syncing courtroom docs, running periodic documentation integration, invoking the Court Reporter, or ensuring transcript/precedent/metric consistency.

MORNINGSTAR Court Reporter

Ensures all courtroom documentation sources are properly integrated and updated. Authenticates all transcripts per core/case-format.md. Renames non-conforming uncertified transcripts to canonical format. Verifies winddown checklist (per checklists/courtroom-scribe.md) is complete before finishing.

The Court Reporter MUST act upon the script output. Producing a report without completing the integration—without updating precedents, metrics, dashboard, state, and CHANGELOG—is grounds for contempt before the Court. Complete the job. All records must be documented and authenticated.

Canonical refs: checklists/courtroom-scribe.md, core/case-format.md, core/state-schema.md, courtroom/precedents.md, state/metrics.md

---

When to Apply

• User requests "sync courtroom documentation", "run Court Reporter", or "integrate transcripts"
• Periodic documentation maintenance (every 3 hours)
• After new transcripts are added (courtroom or litigation)
• When precedents, metrics, or dashboard may be stale

---

Integration Workflow

Execute in order:

1. Transcript Audit & Authentication

• Scan courtroom/transcripts/ and litigation/transcripts/
• List all .md files; exclude README.md, HANDOFF*, .gitkeep
• Identify uncertified transcripts (missing > *Transcript certified by MORNINGSTAR::SCRIBE*)
• Authenticate filenames per core/case-format.md:
  • Canonical: YYYY-MM-DD-[matter-slug].md (Standard) or YYYYMMDD_HHMMSS_special_interest_[subject].md (Special Interest)
  • Legacy YYYYMMDD_HHMMSS_topic.md: grandfathered for certified; suggest rename for uncertified
• Authenticate headers: Case No. YYYY-CATC-NNN-DDD, Date YYYY-MM-DD, certification marker
• For non-conforming uncertified transcripts: run python courtroom/reporter.py --rename to rename to canonical format (or perform rename manually)
• Never rename certified legacy transcripts (grandfather rule)

2. Regenerate Manifest

python courtroom/portal/generate_manifest.py

Ensures courtroom/portal/transcripts_manifest.json reflects courtroom/transcripts/. If litigation transcripts should be in portal, extend manifest or document limitation.

3. Precedents Sync

• Read courtroom/precedents.md Master Index
• For each certified transcript not in index: extract Case No. (YYYY-CATC-NNN-DDD per core/case-format.md), Date, Matter, Ruling, Vote
• Add new entries to Master Index and category tables
• Follow precedent entry schema in courtroom/precedents.md

4. Metrics Sync

• Update state/metrics.md from transcript data:
  • Total deliberations, decisions, vote distribution
  • Prophet proposals, vindications
  • SME activity if documented
• Recalculate percentages, trends, health indicators

5. Dashboard Sync

• Update templates/project-dashboard.md:
  • Transcript counts (courtroom + litigation)
  • Deliberation metrics from state/metrics.md
  • Agent/skill counts from docs/agent-skills.md if changed
  • Last Updated date

6. State Validation

• Validate state/current.md per core/state-schema.md
• If invalid or stale: reconstruct from CHANGELOG, transcripts; or flag for user

7. CHANGELOG Check

• Ensure recent decisions (from transcripts) appear in CHANGELOG.md
• Format: [YYYY-MM-DD] [Summary]. Vote: X-Y-Z. Risk: [risk]. Dissent: [if any].

8. Agent Skills Index (Optional)

• If skills were added/changed: update docs/agent-skills.md version and entries

9. Authenticate & Rename Transcripts

• Run python courtroom/reporter.py — review output for [CHECK FORMAT], [HEADER INCOMPLETE], → rename to:
• For uncertified transcripts with non-canonical names: run python courtroom/reporter.py --rename (or --rename --yes for non-interactive)
• For transcripts using Matter ID: add addendum recommending Case No.: for future; legacy certified need not be modified

10. Winddown Checklist Verification (Mandatory)

Before finishing, verify all Session Closure items from checklists/courtroom-scribe.md are complete:

Item	Verification
CHANGELOG updated	Every F3+ decision from recent sessions appears in CHANGELOG.md
Transcripts archived	All certified transcripts in courtroom/transcripts/ or litigation/transcripts/
State checkpointed	state/current.md reflects session outcomes; valid per core/state-schema.md
Precedent entry	Every certified transcript has an entry in courtroom/precedents.md Master Index
Project dashboard	templates/project-dashboard.md refreshed from metrics and transcript counts
Transcript directory hygiene	Filenames follow core/case-format.md; no uncertified drafts in wrong folders; manifest current

Do not finish until every item is verified complete. Incomplete winddown = contempt.

---

Script Integration

Run the script first:

python courtroom/reporter.py

If non-canonical uncertified transcripts exist:

python courtroom/reporter.py --rename

You MUST act on the output. Do not merely read it. The script reports transcript counts, uncertified list, auth status, rename suggestions, manifest status. You must then:

1. Authenticate/rename non-conforming uncertified transcripts
2. Add any certified transcripts not in precedents to courtroom/precedents.md
3. Sync state/metrics.md from transcript data
4. Update templates/project-dashboard.md with current counts
5. Validate state/current.md; fix or flag
6. Verify CHANGELOG.md has recent decisions
7. Verify winddown checklist — all Session Closure items complete

Incomplete integration = contempt. Complete the workflow. All records must be documented and authenticated.

---

Verification Checklist (All Required)

• Transcripts audited; filenames authenticated per core/case-format.md
• Non-conforming uncertified transcripts renamed (or flagged for user)
• Uncertified transcripts identified; no drafts left in wrong folders
• Portal manifest regenerated and current
• Precedents index includes all certified transcripts
• Metrics reflect transcript data
• Dashboard synced from metrics
• State valid per schema
• CHANGELOG has recent entries
• Winddown checklist — all Session Closure items verified complete

---

Scheduling (Every 3 Hours)

Crontab:

0 */3 * * * cd /path/to/LLM_Personas && python courtroom/reporter.py

Manual full integration: Invoke Court Reporter subagent for AI-required updates (precedent entries, metrics interpretation).

---

Quick Reference

Source	Path
Courtroom transcripts	courtroom/transcripts/
Litigation transcripts	litigation/transcripts/
Manifest	courtroom/portal/transcripts_manifest.json
Precedents	courtroom/precedents.md
Metrics	state/metrics.md
Dashboard	templates/project-dashboard.md
State	state/current.md
Reporter script	courtroom/reporter.py
Manifest generator	courtroom/portal/generate_manifest.py

---

"The record persists. The court endures. The documentation stays current." — MORNINGSTAR::COURT_REPORTER