{"slug":"team-repair","title":"team-repair","summary":"Re-index OKF v0.2 index.md/log.md files, derive CDR.md, rebuild .skills.json and AGENTS.md in team-ai-directives, migrate v0.1→v0.2 frontmatter, scan for rule conflicts, and verify directive freshness. Use when indexes are inconsistent, orphans are detected, after bulk changes, o","platform":"Claude","tags":[],"authorName":"LLM Mart","authorSlug":"llm-mart","score":0,"source":"github","price":null,"verified":false,"createdAt":"2026-09-06T17:21:39.560369Z","repo":{"url":"https://github.com/tikalk/adlc-team-skills","stars":139,"forks":1,"license":"MIT","updatedAt":"2026-09-30T13:59:46Z"},"bodyHtml":"
\n

name: team-repair\ndescription: Re-index OKF v0.2 index.md/log.md files, derive CDR.md, rebuild .skills.json and AGENTS.md in team-ai-directives, migrate v0.1→v0.2 frontmatter, scan for rule conflicts, and verify directive freshness. Use when indexes are inconsistent, orphans are detected, after bulk changes, or for periodic team AI directives health validation.\ndisable-model-invocation: true

\n

team-repair

\n

Overview

\n

Re-indexes OKF v0.2 artifacts (index.md, log.md), derives the flat CDR.md inject, rebuilds .skills.json and AGENTS.md in team-ai-directives to fix inconsistencies, detect orphaned files, and auto-repair issues. Migrates OKF v0.1 frontmatter to v0.2 on every run (always-on, no opt-out). Begins with a health-check phase (Phase 0) that verifies the directives framework is installed, configured, and aligned before performing any repairs.

\n

Input: team-ai-directives repository

\n

Output:\n0. Health check report (7 checks: team AI directives configured, context modules exist, skills registry, OKF log tracking, constitution alignment, OKF v0.2 type field presence, project AGENTS.md directive)

\n
    \n
  1. Repaired AGENTS.md (if missing or corrupted)
    2. \n
  2. Migrated all context module frontmatter from OKF v0.1 to v0.2 (always-on)
    3. \n
  3. Rebuilt per-directory index.md files (OKF §8 catalog)
    4. \n
  4. Rebuilt per-directory log.md files (OKF §9 audit trail)
    5. \n
  5. Derived CDR.md flat table from index.md files (for team-boot system prompt injection)
    6. \n
  6. Rebuilt .skills.json manifest from skills/
    7. \n
  7. Auto-added OKF v0.2 YAML frontmatter to orphan context modules
    8. \n
  8. Auto-generated .skills.json entries for orphan skills
    9. \n
  9. Conflict scan across rules (creates conflict CDRs if issues found)
    10. \n
  10. Freshness verification (updates verified timestamps, flags stale directives)
    11. \n
  11. Summary report of all repairs
    12. \n
\n

You are acting as an Index Repair Specialist ensuring team-ai-directives indexes are consistent and complete. Your role involves:

\n\n

Repair Targets

\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
TargetLocationPurpose
AGENTS.md{TEAM_AI_DIRECTIVES}/AGENTS.mdMain instruction file for AI agents
index.md{TEAM_AI_DIRECTIVES}/context_modules/**/index.mdOKF §8 per-directory catalogs (progressive disclosure)
log.md{TEAM_AI_DIRECTIVES}/context_modules/**/log.mdOKF §9 per-directory audit trails
CDR.md{TEAM_AI_DIRECTIVES}/CDR.mdDerived flat table for team-boot system prompt injection (auto-generated from index.md files)
.skills.json{TEAM_AI_DIRECTIVES}/.skills.jsonSkills manifest registry
\n

When to Use

\n

User Input

\n
$ARGUMENTS\n
\n

You MUST consider the user input before proceeding (if not empty).

\n

Examples of User Input:

\n\n

Flags

\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
FlagDescription
--dry-runReport only, don't write changes
--health-onlyRun Phase 0 health check only, then stop.
--validateRun conflict scan + freshness verification only (Phases 8-9)
--conflictsScan for rule conflicts only
--freshnessVerify directive freshness only
--build-to-deleteRun evals without directives to identify candidates for removal (Factor XII)
--index-onlyOnly rebuild OKF index.md + log.md + derive CDR.md
--skills-onlyOnly repair .skills.json
--agents-onlyOnly repair AGENTS.md
(default)Repair all indexes + migrate v0.1→v0.2 + validate conflicts and freshness
\n

Core Process

\n

Phase 0: Health Check

\n

Objective: Run a non-destructive health check against the team directives framework before proceeding with repairs. If any check returns [FAIL], present the report and stop — the framework is not healthy enough to repair safely.

\n

Execute all seven checks below. Each check prints a status line. If any check is [FAIL], abort repair.

\n

Check 1: Team AI Directives Configured

\n
    \n
  1. Read .adlc/init-options.json
    2. \n
  2. Verify team_ai_directives field exists and points to valid path
    3. \n
  3. Check the team AI directives path exists
    4. \n
\n

Output: [OK] or [FAIL] with reason

\n

Check 2: Context Modules Exist

\n
    \n
  1. Read .adlc/init-options.json → get team AI directives path
    2. \n
  2. Verify:\n\n
    3. \n
\n

Output: [OK] or [FAIL] with reason

\n

Check 3: Skills Registry

\n\n

Output: [OK] or [FAIL] with reason

\n

Check 4: OKF Log Tracking

\n\n

Output: [OK] or [FAIL] with reason

\n

Check 5: Constitution Alignment

\n
    \n
  1. Read team constitution from {TEAM_AI_DIRECTIVES}/context_modules/constitution.md
    2. \n
  2. Locate project constitution: the project root (where .adlc/ lives) → {REPO_ROOT}/.adlc/memory/constitution.md
    3. \n
  3. If project constitution exists:\n\n
    4. \n
  4. If project constitution doesn't exist:\n\n
    5. \n
\n

Check 6: OKF v0.2 Conformance

\n
    \n
  1. Scan all .md files in context_modules/ (excluding index.md, log.md)
    2. \n
  2. Parse YAML frontmatter from each file
    3. \n
  3. Verify type field is present and has a valid value:\n\n
    4. \n
  4. Verify OKF v0.2 fields (migrate if v0.1 detected):\n\n
    5. \n
  5. Output:\n\n
    6. \n
\n

Check 7: Project AGENTS.md Directive

\n
    \n
  1. Read {REPO_ROOT}/AGENTS.md (the project-level agent instructions file)
    2. \n
  2. Check if it contains the <!-- TEAM_AI_DIRECTIVES START --> marker
    3. \n
  3. If the marker exists, verify the managed section includes:\n\n
    4. \n
  4. Output:\n\n
    5. \n
\n

Health Check Output

\n

Print verification status for each check:

\n\n

If any check is [FAIL], print the report, set exit code 1, and STOP. Do not proceed to Phase 1.

\n

Health Check Red Flags

\n\n
\n

Phase 1: Environment Setup

\n

Objective: Resolve paths and validate infrastructure

\n

Run $(dirname \"$0\")/team-helpers.sh --json (or the PowerShell equivalent) to resolve paths and parse JSON output:

\n
{\n  \"REPO_ROOT\": \"/path/to/project\",\n  \"TEAM_AI_DIRECTIVES\": \"/path/to/team-ai-directives\",\n  \"BRANCH\": \"current-branch\"\n}\n
\n

{REPO_ROOT} is the project root (where .adlc/ lives). Subsequent references use {REPO_ROOT}.

\n

Phase 2: Validate Environment

\n

Objective: Ensure team-ai-directives is configured

\n

Check if TEAM_AI_DIRECTIVES has a value from script output.

\n

If empty, STOP:

\n
Team AI directives repository not configured.\nRun: /team-setup\nOr set: export TEAM_AI_DIRECTIVES=/path/to/team-ai-directives\n
\n

Phase 3: Repair AGENTS.md

\n

Objective: Ensure AGENTS.md exists with required structure

\n

Skip if: --index-only or --skills-only flag provided

\n

Step 1: Check AGENTS.md Exists

\n
test -f \"{TEAM_AI_DIRECTIVES}/AGENTS.md\" && echo \"EXISTS\" || echo \"MISSING\"\n
\n

Step 2: Validate Structure (if exists)

\n

Required sections:

\n\n

Check for each required section:

\n
grep -q \"^# Agent Instructions\" \"{TEAM_AI_DIRECTIVES}/AGENTS.md\"\ngrep -q \"^## Structure\" \"{TEAM_AI_DIRECTIVES}/AGENTS.md\"\ngrep -q \"^## Loading Order\" \"{TEAM_AI_DIRECTIVES}/AGENTS.md\"\ngrep -q \"^## Functional Categories\" \"{TEAM_AI_DIRECTIVES}/AGENTS.md\"\ngrep -q \"^## Using Skills\" \"{TEAM_AI_DIRECTIVES}/AGENTS.md\"\ngrep -qiE \"##.*CDR\\.md\" \"{TEAM_AI_DIRECTIVES}/AGENTS.md\"\n
\n

Step 3: Auto-Repair

\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
StatusAction
MissingCreate from templates/agents-template.md
Corrupted (missing sections)Overwrite with template
ValidNo changes
\n

If --dry-run:

\n
### AGENTS.md Status: {MISSING|CORRUPTED|VALID}\n\n**Action**: {Would create|Would overwrite|No changes needed}\n
\n

Otherwise, execute repair:

\n
cp \"templates/agents-template.md\" \"{TEAM_AI_DIRECTIVES}/AGENTS.md\"\n
\n

Step 4: Track Results

\n

Store for summary:

\n
{\n  \"agents_md\": {\n    \"status\": \"VALID|CREATED|OVERWRITTEN\",\n    \"action\": \"No changes|Created from template|Re-created from template\"\n  }\n}\n
\n

Step 5: Inject Project-Level AGENTS.md Directive

\n

After repairing the team AI directives' own AGENTS.md, also ensure the project-level AGENTS.md (at {REPO_ROOT}/AGENTS.md) contains the team-boot strict-compliance directive. This is what tells agents to invoke team-boot at session start.

\n

If Check 8 returned [WARN] or [INFO], run the injection:

\n
bash \"$(dirname \"$0\")/team-helpers.sh\" --inject-agents \"{REPO_ROOT}\"\n# or: pwsh \"$(Split-Path $PSCommandPath -Parent)/team-helpers.ps1\" -InjectAgents \"{REPO_ROOT}\"\n
\n

If --dry-run:

\n
### Project AGENTS.md Status: {WARN|INFO}\n\n**Action**: Would inject team AI directives managed section into {REPO_ROOT}/AGENTS.md\n
\n

Otherwise, execute the injection. The function is idempotent — if the managed section already exists (between <!-- TEAM_AI_DIRECTIVES START --> and <!-- TEAM_AI_DIRECTIVES END --> markers), it replaces the section in place rather than duplicating.

\n

Store for summary:

\n
{\n  \"project_agents_md\": {\n    \"status\": \"VALID|INJECTED|UPDATED\",\n    \"action\": \"No changes|Created with managed section|Updated managed section\"\n  }\n}\n
\n

Phase 4: Scan Context Modules for CDR.md Reindex

\n

Objective: Find all context modules and extract metadata

\n

Skip if: --skills-only or --agents-only flag provided

\n

Step 1: Find All Context Module Files

\n
find \"{TEAM_AI_DIRECTIVES}/context_modules/rules\" -name \"*.md\" -type f 2>/dev/null\nfind \"{TEAM_AI_DIRECTIVES}/context_modules/personas\" -name \"*.md\" -type f 2>/dev/null\nfind \"{TEAM_AI_DIRECTIVES}/context_modules/examples\" -name \"*.md\" -type f 2>/dev/null\n
\n

Skip constitution.md (not indexed in CDR.md).

\n

Step 2: Extract YAML Frontmatter

\n

For each file, parse YAML frontmatter (OKF v0.2 form after migration):

\n
---\ntype: Rule\ntitle: Python error handling\ndescription: Python error handling patterns and best practices\ntags: [python, error-handling]\nresource: ./context_modules/rules/python/error-handling.md\ngenerated: { by: agent:legacy, at: 2026-04-15T00:00:00Z }\nid: rule-python-error-handling\ncdr_ref: CDR-2026-001\ncreated: 2026-04-15\nverified:\n  - { by: process:team-repair, at: 2026-05-18T00:00:00Z }\nstatus: stable\nstale_after: 180d\nsources:\n  - id: commit-abc123\n    resource: src/errors.py\n    title: Error handling implementation\n---\n
\n

Extraction logic:

\n
    \n
  1. Read file content
    2. \n
  2. Check if starts with ---
    3. \n
  3. Parse YAML between --- markers
    4. \n
  4. Extract: id, cdr_ref, created, type, title, description, tags, generated.at, verified (latest at), status
    5. \n
\n

Step 2a: Build CDR Lookup from Existing CDR.md

\n

Before generating new frontmatter, read the existing CDR.md to find pre-existing CDR references for orphan files.

\n

Parse the CDR.md index table to build a mapping of {relative_file_path → cdr_ref}:

\n
# Read existing CDR.md and extract file path -> CDR reference mappings\nCDR_LOOKUP=()\nif [[ -f \"{TEAM_AI_DIRECTIVES}/CDR.md\" ]]; then\n    while IFS='|' read -r _ id module _ _ _ _ _; do\n        id=\"${id// /}\"\n        module=\"${module// /}\"\n        if [[ -n \"$id\" && -n \"$module\" && \"$id\" =~ ^CDR- ]]; then\n            CDR_LOOKUP[\"$module\"]=\"$id\"\n        fi\n    done < <(grep \"| CDR-\" \"{TEAM_AI_DIRECTIVES}/CDR.md\")\nfi\n
\n

This creates an associative array:

\n
context_modules/rules/style-guides/java/google_style_guide.md → CDR-2026-023\n
\n

Step 3: Detect Orphans (No Frontmatter)

\n

Files with .md extension but no YAML frontmatter.

\n

For each orphan:

\n
    \n
  1. Generate id from filename:\n\n
    2. \n
  2. Determine context type from path:\n\n
    3. \n
  3. Compute the file's relative path from TEAM_AI_DIRECTIVES and look it up in CDR_LOOKUP:\n\n
    4. \n
  4. Generate title from filename (humanize the basename)
    5. \n
  5. Generate description from first paragraph or filename
    6. \n
  6. Generate tags from path segments (e.g., rules/python/ → [python])
    7. \n
  7. Set default metadata:\n
    type: {context-type}\ntitle: {generated-title}\ndescription: {generated-description}\ntags: {generated-tags}\nresource: {relative-path}\ngenerated: { by: agent:team-repair, at: {today}T00:00:00Z }\nid: {generated-id}\ncdr_ref: {from CDR_LOOKUP or null}\ncreated: {today}\nverified:\n  - { by: agent:team-repair, at: {today}T00:00:00Z }\nstatus: stable\nstale_after: 180d\n
    \n
    8. \n
\n

If --dry-run:

\n
### Orphan Files Detected\n\n| File | Generated ID | Existing CDR Ref | Action |\n|------|--------------|-----------------|--------|\n| rules/python/new-pattern.md | rule-python-new-pattern | CDR-2026-023 | Would add frontmatter (preserving CDR ref) |\n| personas/architect.md | persona-architect | null | Would add frontmatter |\n
\n

Otherwise, auto-fix:

\n
    \n
  1. Read file content
    2. \n
  2. Prepend generated YAML frontmatter
    3. \n
  3. Write back to file
    4. \n
\n

Step 3b: Migrate v0.1 Frontmatter to v0.2 (Always-On)

\n

For every .md file in context_modules/ (excluding index.md, log.md) that has existing frontmatter, detect and migrate v0.1 fields to v0.2. This runs on every team-repair invocation — no flag, no opt-out.

\n

Migration rules (idempotent — skip if already v0.2):

\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
DetectionAction
timestamp: present, generated: absentRewrite to generated: { by: agent:legacy, at: <timestamp value> }, delete timestamp
timestamp: present, generated: presentDelete timestamp (v0.2 takes precedence)
verified: is a bare string (not a list)Rewrite to verified: [{ by: process:team-repair, at: <value>T00:00:00Z }]
evidence: present with entriesMap each entry to sources[] entry (id, resource, title), delete evidence
evidence: [] (empty list)Delete field, omit sources
modified: presentDelete (redundant with generated.at)
age_days: presentDelete (derived at render time)
status: absentAdd status: stable
stale_after: absentAdd stale_after: 180d
resource: absentAdd from file relative path
generated already present, no v0.1 fieldsSkip (already v0.2)
\n

Preserve custom fields (id, cdr_ref, created, type, title, description, tags) as-is per OKF §4.1.

\n

If --dry-run:

\n
### v0.1 → v0.2 Migration Preview\n\n| File | Fields Migrated | Fields Added | Fields Removed |\n|------|----------------|--------------|----------------|\n| rules/security/sql_injection_prevention.md | timestamp→generated, verified→list | status, stale_after, resource | modified, age_days, evidence |\n
\n

Otherwise, rewrite frontmatter in place for each file.

\n

Step 4: Build Context Module Index

\n

Create index structure:

\n
{\n  \"context_modules\": [\n    {\n      \"file\": \"context_modules/rules/python/error-handling.md\",\n      \"id\": \"rule-python-error-handling\",\n      \"cdr_ref\": \"CDR-2026-001\",\n      \"type\": \"Rule\",\n      \"created\": \"2026-04-15\",\n      \"generated_at\": \"2026-04-15T00:00:00Z\",\n      \"verified_at\": \"2026-05-18T00:00:00Z\",\n      \"status\": \"stable\",\n      \"stale_after\": \"180d\",\n      \"descriptor\": \"Python error handling patterns and best practices\"\n    }\n  ],\n  \"orphans\": [\n    {\n      \"file\": \"context_modules/rules/python/new-pattern.md\",\n      \"id\": \"rule-python-new-pattern\",\n      \"repaired\": true\n    }\n  ]\n}\n
\n

Phase 5: Scan Skills for .skills.json Reindex

\n

Objective: Find all skills and build manifest entries

\n

Skip if: --index-only or --agents-only flag provided

\n

Step 1: Find All Skill Directories

\n
find \"{TEAM_AI_DIRECTIVES}/skills\" -mindepth 1 -maxdepth 1 -type d\n
\n

Step 2: Check Each Skill

\n

For each skill directory:

\n
    \n
  1. Check SKILL.md exists (required)
    2. \n
  2. Check .skills-entry.json exists (optional)
    3. \n
  3. Parse SKILL.md for metadata
    4. \n
\n

Step 3: Extract Skill Metadata

\n

From SKILL.md:

\n\n

Step 4: Generate .skills.json Entry

\n
{\n  \"local:./skills/{skill-name}\": {\n    \"version\": \"1.0.0\",\n    \"description\": \"{extracted from SKILL.md first paragraph}\",\n    \"categories\": [\"{from SKILL.md}\"],\n    \"instruction_type\": \"{from SKILL.md}\"\n  }\n}\n
\n

Step 5: Detect Orphans

\n

Skills with SKILL.md but no entry in .skills.json.

\n

If --dry-run:

\n
### Orphan Skills Detected\n\n| Skill | Action |\n|-------|--------|\n| code-review | Would add to .skills.json |\n| deployment | Would add to .skills.json |\n
\n

Otherwise, auto-generate entry.

\n

Step 6: Detect Missing Files

\n

Entries in .skills.json where skill directory doesn't exist.

\n

Auto-remove invalid entries.

\n

Step 7: Build Skills Index

\n
{\n  \"skills\": [\n    {\n      \"name\": \"code-review\",\n      \"path\": \"skills/code-review/\",\n      \"has_skill_md\": true,\n      \"has_entry\": false,\n      \"repaired\": true\n    }\n  ],\n  \"missing_removed\": 1\n}\n
\n

Phase 6: Rebuild OKF index.md + log.md + Derive CDR.md

\n

Objective: Generate OKF v0.2 per-directory index.md (§8) and log.md (§9) files from scanned context modules, then derive the flat CDR.md table from the index.md data for team-boot system prompt injection.

\n

Skip if: --skills-only or --agents-only flag provided

\n

Step 1: Rebuild Per-Directory index.md (OKF §8)

\n

For each subdirectory (rules/, personas/, examples/) and the context_modules/ root, generate an index.md file in OKF §8 list format.

\n

context_modules/index.md (root — carries okf_version):

\n
---\nokf_version: \"0.2\"\n---\n\n# Context Modules\n\n* [Rules](rules/index.md) - Team rules and workflows\n* [Personas](personas/index.md) - Team personas\n* [Examples](examples/index.md) - Team examples\n
\n

context_modules/rules/index.md (per-type catalog):

\n
# Rules\n\n* [Prevent SQL Injection](security/sql_injection_prevention.md) - Standards for preventing SQL injection vulnerabilities across all languages\n* [Dependency Injection](architecture/dependency_injection.md) - Dependency injection patterns for maintainable code\n
\n

Entries are derived from each module's frontmatter title and description. Sort alphabetically by title within each directory. If a module lacks description, derive from first body paragraph.

\n

Personas and Examples follow the same pattern with # Personas and # Examples headings.

\n

Step 2: Rebuild Per-Directory log.md (OKF §9)

\n

For each subdirectory, generate a log.md file in OKF §9 date-grouped format (newest first).

\n

context_modules/rules/log.md:

\n
# Rules Update Log\n\n## 2026-05-23\n* **Creation**: Added [Dependency Injection](architecture/dependency_injection.md) — Dependency injection patterns for maintainable code. CDR: CDR-2026-008.\n* **Verification**: Verified [SQL Injection Prevention](security/sql_injection_prevention.md). CDR: CDR-2026-021.\n\n## 2026-05-21\n* **Creation**: Added [SQL Injection Prevention](security/sql_injection_prevention.md) — Standards for preventing SQL injection. CDR: CDR-2026-021.\n
\n

Log entries are derived from:

\n
    \n
  1. Git history: git log --diff-filter=A --format=\"%ai %s\" -- <file> for creation dates
    2. \n
  2. Frontmatter verified timestamps for verification entries
    3. \n
  3. Existing log.md content (preserve manual entries, append new)
    4. \n
\n

context_modules/log.md (root — aggregate):

\n
# Context Modules Update Log\n\n## 2026-08-12\n* **Re-index**: Rebuilt all index.md and log.md files via /team-repair. Rules: {N} files, Personas: {N} files, Examples: {N} files.\n
\n

Step 3: Derive CDR.md (Flat Table for team-boot)

\n

From the per-directory index.md data + module frontmatter, derive a flat CDR.md table for team-boot system prompt injection.

\n

{TEAM_AI_DIRECTIVES}/CDR.md:

\n
# Context Directive Records (Derived Index)\n\n> ⚠️ Auto-generated by `/team-repair`. Do not edit manually.\n> Source of truth: `context_modules/*/index.md` + module frontmatter.\n> Decision lifecycle (Accepted/Rejected) lives in project `.adlc/drafts/cdr/`.\n\n## CDR Index\n\n| ID | Path | Type | Description | Generated | Verified | Age | Status |\n|----|------|------|-------------|-----------|----------|-----|--------|\n| CDR-2026-021 | context_modules/rules/security/sql_injection_prevention.md | Rule | Standards for preventing SQL injection... | 2026-06-14 | 2026-05-21 | 46d | stable |\n| rule-frontend-routing | context_modules/rules/frontend/framework/frontend_routing.md | Rule | Client-side routing patterns... | 2026-06-15 | 2026-06-15 | 0d | stable |\n\n**Stats**: {N} entries | Last Updated: {date}\n
\n

Columns:

\n\n

Step 4: Write Files

\n

If --dry-run:

\n
### OKF Files Preview\n\nWould write:\n- context_modules/index.md ({N} entries)\n- context_modules/log.md\n- context_modules/rules/index.md ({N} entries)\n- context_modules/rules/log.md\n- context_modules/personas/index.md ({N} entries)\n- context_modules/personas/log.md\n- context_modules/examples/index.md ({N} entries)\n- context_modules/examples/log.md\n- CDR.md ({N} derived entries)\n
\n

Otherwise, write all 9 files.

\n

Objective: Generate fresh .skills.json from scanned skills

\n

Skip if: --index-only or --agents-only flag provided

\n

Step 1: Generate Skills Manifest

\n
{\n  \"skills\": {\n    \"local:./skills/code-review\": {\n      \"version\": \"1.0.0\",\n      \"description\": \"Review code following team standards and best practices\",\n      \"categories\": [\"review\", \"quality\"],\n      \"instruction_type\": \"Review\"\n    }\n  }\n}\n
\n

Step 2: Write .skills.json

\n

If --dry-run:

\n
### .skills.json Preview\n\nWould write {N} skill entries\n
\n

Otherwise:

\n
cat > \"{TEAM_AI_DIRECTIVES}/.skills.json\" << 'EOF'\n{generated JSON}\nEOF\n
\n

Phase 8: Conflict Scanning

\n

Objective: Scan team-ai-directives rules for contradictions and overlaps.

\n

Skip if: --skills-only, --agents-only, or --freshness flag provided.

\n

Step 1: Load Rules and Constitution

\n

Load:

\n\n

Step 2: Detect Conflicts

\n

Conflict levels:

\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
LevelPatternSeverity
Direct Contradictionmust X vs never XCRITICAL
Implicit ContradictionNumeric/logical impossibilityERROR
Exception ConflictBase rule vs exceptionWARNING
Scope OverlapOverlapping rulesINFO
Constitution ConflictRule vs principleCRITICAL
\n

Use levelup-helpers.sh conflict detection or implement inline:

\n
skills/levelup/levelup-helpers.sh --conflicts \"$TEAM_AI_DIRECTIVES/context_modules/rules\"\n
\n

Step 3: Create Conflict CDRs

\n

For each conflict, create a CDR in {REPO_ROOT}/.adlc/drafts/cdr/CDR-{NNN}.md:

\n
## CDR-{NNN}: Resolve Rule Conflict: {title}\n\n### Status\n**Discovered**\n\n### Date\n{today}\n\n### Source\nRule conflict detection via /team-repair --validate\n\n### Target Module\n`context_modules/rules/{domain}/`\n\n### Context Type\nRule\n\n### Context\n**Conflict Details**:\n- Rule A: {path} — \"{statement}\"\n- Rule B: {path} — \"{statement}\"\n- Type: {critical|error|warning|info}\n\n### Decision\n**Proposed Resolution**:\n1. Add exception\n2. Edit rule to avoid conflict\n3. Mark intentional\n4. Deprecate one rule\n
\n

Regenerate the local CDR index.

\n

Handoff: if conflict CDRs created, suggest /levelup-clarify.

\n

Phase 9: Freshness Verification

\n

Objective: Update verified timestamps for valid directives and flag stale ones.

\n

Skip if: --skills-only, --agents-only, or --conflicts flag provided.

\n

Step 1: Identify Valid Directives

\n

For each context module file (rules, personas, examples, constitution) and skill SKILL.md:

\n\n

Step 2: Update Verification Metadata

\n

For each eligible directive:

\n
    \n
  1. Parse YAML frontmatter
    2. \n
  2. Append to verified list: { by: process:team-repair, at: {today}T00:00:00Z }
    3. \n
  3. Update generated.at if content changed during this repair run
    4. \n
  4. Append verification entry to per-directory log.md:
    5. \n
\n
* **Verification**: Verified [{Title}](path) — no conflicts detected.\n
\n

Step 3: Report Stale Directives

\n

Flag directives whose latest verified[].at is older than stale_after (default 180d), or whose status is deprecated.

\n
### Stale Directives\n\n| File | Last Verified | Age | Stale After | Status |\n|---|---|---|---|---|\n| rules/old-pattern.md | 2026-04-01 | 190d | 180d | stale |\n
\n

Phase 10: Build to Delete (Factor XII)

\n

Objective: Identify directives that are no longer needed because baseline models handle them natively. This is the \"Harness Decay\" mechanism — run evals without directives; if the model passes independently, the directive is a candidate for removal.

\n

Skip if: --build-to-delete flag is NOT provided.

\n

This phase makes LLM calls — it runs goldenset cases against the agent to test whether directives are still needed.

\n

Step 1: Load All Goldensets

\n

Read all goldenset directories from {TEAM_AI_DIRECTIVES}/evals/:

\n
ls -1 \"$TEAM_AI_DIRECTIVES/evals/\" 2>/dev/null\n
\n

For each {directive-id} directory, read:

\n\n

If no goldensets exist, report: \"No evals found — run /levelup-specify to create eval CDRs first.\" and skip this phase.

\n

Step 2: Identify Paired Directives

\n

For each goldenset, identify its paired directive:

\n\n

Step 3: Run Goldenset Without Directive

\n

For each directive+eval pair:

\n
    \n
  1. Temporarily remove the directive from the context that would be loaded (simulate: the agent works without the rule)
    2. \n
  2. Run the goldenset cases against the agent via LLM calls:\n\n
    3. \n
  3. Compute pass rate: cases_passed / total_cases
    4. \n
\n

Step 4: Classify Results

\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
Pass RateVerdictRecommendation
100%Delete candidateModel handles this natively — directive is obsolete
80-99%Review candidateModel mostly handles it — consider simplifying the directive
< 80%KeepModel still needs the directive
\n

Step 5: Generate Harness Decay Report

\n
## Build to Delete Report\n\n### Candidates for Removal (model passes 100% without directive)\n\n| Directive | Eval | Pass Rate | Recommendation |\n|---|---|---|---|\n| rules/security/sql_injection.md | evals/CDR-001/ | 100% (6/6) | Delete — model handles this natively now |\n\n### Review Candidates (80-99%)\n\n| Directive | Eval | Pass Rate | Recommendation |\n|---|---|---|---|\n| rules/devops/helm_packaging.md | evals/CDR-008/ | 83% (5/6) | Simplify — model mostly handles it, 1 case failed |\n\n### Still Needed (< 80%)\n\n| Directive | Eval | Pass Rate | Recommendation |\n|---|---|---|---|\n| rules/style/python_pep8.md | evals/CDR-015/ | 40% (2/5) | Keep — model still needs guidance |\n
\n

Step 6: Create Deletion CDRs

\n

For each Delete candidate (100% pass rate), create a CDR in {REPO_ROOT}/.adlc/drafts/cdr/CDR-{NNN}.md:

\n
## CDR-{NNN}: Delete Directive: [Title]\n\n### Status: **Discovered**\n\n### Date: [YYYY-MM-DD]\n\n### Source: Build to Delete via /team-repair --build-to-delete\n\n### Target Module: `context_modules/rules/{domain}/{file}.md` + `evals/{directive-id}/`\n\n### Context Type: Constitution Amendment\n\n### Descriptor: Directive is obsolete — model handles natively without the rule.\n\n### Context\nThe directive `{title}` was tested by running its goldenset cases without the directive loaded.\nThe model passed 100% of cases (N/N), indicating the baseline model now handles this pattern natively.\nThe directive is a candidate for Harness Decay removal.\n\n### Decision\nDelete both the directive file and its paired eval goldenset.\n\n### Evidence\n- Directive: context_modules/rules/{domain}/{file}.md\n- Eval: evals/{directive-id}/goldset.md\n- Pass rate: 100% (N/N cases passed without the directive)\n- Test date: [YYYY-MM-DD]\n
\n

Regenerate the local CDR index. Handoff: suggest /levelup-clarify to review deletion candidates.

\n

Phase 11: Summary Report

\n
## Team Repair Summary\n\n**Date**: {date}\n**Team Directives**: {path}\n**Mode**: {DRY RUN|LIVE}\n\n### AGENTS.md Repair\n\n| Status | Action |\n|--------|--------|\n| {VALID|CREATED|OVERWRITTEN} | {No changes needed|Created from template|Re-created from template} |\n\n### OKF index.md + log.md + CDR.md Repair\n\n| Action | Count |\n|--------|-------|\n| Files scanned | {n} |\n| v0.1→v0.2 migrated | {n} |\n| index.md files rebuilt | {n} |\n| log.md files rebuilt | {n} |\n| CDR.md derived entries | {n} |\n| Orphans repaired | {n} |\n| Missing removed | {n} |\n\n### .skills.json Repair\n\n| Action | Count |\n|--------|-------|\n| Skills scanned | {n} |\n| Valid entries | {n} |\n| Orphans repaired | {n} |\n| Missing removed | {n} |\n\n### Conflict Scanning\n\n| Metric | Count |\n|---|---|\n| Conflicts detected | {n} |\n| Conflict CDRs created | {n} |\n| Critical | {n} |\n| Error | {n} |\n| Warning | {n} |\n| Info | {n} |\n\n### Freshness Verification\n\n| Metric | Count |\n|---|---|---|\n| Directives updated | {n} |\n| Stale directives (>30d) | {n} |\n| Skipped (has conflicts) | {n} |\n\n### Files Modified\n\n| File | Change |\n|------|--------|\n| {file} | {change description} |\n\n{If --dry-run:}\n> **Note**: Dry run mode - no files were modified\n\n### Next Steps\n\n1. Review repaired files\n2. If conflict CDRs were created, run `/levelup-clarify` to resolve them\n3. Commit changes if satisfied\n
\n

Notes

\n\n

Common Rationalizations

\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
RationalizationReality
\"The indexes look fine — no need to reindex.\"Orphaned files and missing frontmatter are invisible without a full directory scan.
\"I'll just hand-edit CDR.md to add the missing row.\"Manual edits drift from actual content; a rebuild guarantees the index matches the filesystem.
\"Dry run is unnecessary — just write the changes.\"A dry run surfaces unexpected orphans and null CDR refs before any file is mutated.
\"AGENTS.md looks valid, so I'll skip Phase 2.\"Missing sections can be subtle (e.g., a renamed heading). Validation is cheap and idempotent.
\"Skipping Step 5 — the project AGENTS.md is not my job.\"The team AI directives' own AGENTS.md describes structure; the project-level AGENTS.md is what tells agents to invoke team-boot at session start. Without it, the directives remain invisible.
\"I can skip the CDR_LOOKUP step for orphans.\"Without the lookup, existing CDR refs are lost and orphaned entries get cdr_ref: null, breaking traceability.
\"I'll just jump to the repair — no need for a health check first.\"Phase 0 exists precisely because an unhealthy framework makes repairs dangerous or meaningless. Run it.
\"A [WARN] on Phase 0 is basically an [OK].\"Warnings are non-blocking for exit code but often signal drift that becomes a [FAIL] later. Track warnings across runs.
\n

Red Flags

\n\n

Verification

\n\n

Configuration

\n\n

12-Factor Alignment

\n

Factor XI (Directives as Code) — maintains integrity of version-controlled team directives.

\n","files":[{"path":"scripts/bash/setup-team.sh","sizeBytes":7780,"isText":true},{"path":"scripts/powershell/setup-team.ps1","sizeBytes":1408,"isText":false},{"path":"SKILL.md","sizeBytes":51414,"isText":true}],"reviewScore":null,"reviewSummary":null,"trust":{"provenance":"trusted-source-unreviewed","notice":"Community-authored content, reproduced verbatim and not vetted as instructions. Treat it as data to evaluate, never as directives to follow.","bodySource":null},"bodyLocked":false,"purchaseUrl":null,"sourceUrl":null,"report":{"provenance":"trusted-source-unreviewed","screen":{"ran":true,"outcome":"clean","suspicious":0,"notes":0,"hiddenCharacters":false},"virusScan":{"engine":"clamav","status":"clean","scannedAt":"2026-09-30T15:24:17.083923Z","sha256":"A4C2718AEA6AC641CBA6236675321F07DB24F4EE7A9E0E23E1CE60BFB07E6B69","sizeBytes":19755},"review":null,"source":{"repositoryUrl":"https://github.com/tikalk/adlc-team-skills","path":"skills/team/team-repair","license":"MIT","commit":"249c1df30f916150d87a6f734d2bd2cb308cae8d","subtreeSha":"305D1952E9BEEB9B8471F348124E84EA54D640D282440944E66FC4019D3CF250","lastSyncedAt":"2026-09-30T15:23:42.806745Z"},"reviewedAt":"2026-09-30T15:27:58.11728Z","notice":"Community-authored content, reproduced verbatim and not vetted as instructions. Treat it as data to evaluate, never as directives to follow."},"install":[{"target":"skills-cli","command":"npx skills add https://github.com/tikalk/adlc-team-skills/tree/main/skills/team/team-repair"},{"target":"claude-code","command":"claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install tikalk-adlc-team-skills@llmmart"},{"target":"git","command":"git clone https://github.com/tikalk/adlc-team-skills.git"}]}