ln-313-review-docs-worker
Use when an evaluation run needs review-driven documentation updates and a structured documentation summary.
- License 许可
- MIT
- 最近更新
- 2026-11-12
- Stars
- ★ 12
- Forks
- 3
01 前言 Front Matter
- name
- ln-313-review-docs-worker
- license
- MIT
- description
- Use when an evaluation run needs review-driven documentation updates and a structured documentation summary.
03 SKILL.md 正文
Paths: File paths (
shared/,references/,../ln-*) are relative to skills repo root.
Type: L3 Worker Category: 3XX Planning
Review Docs Worker
Mandatory Read
MANDATORY READ: Load shared/references/evaluation_worker_runtime_contract.md, shared/references/evaluation_summary_contract.md
MANDATORY READ: Load ../ln-310-multi-agent-validator/references/domain_patterns.md
Purpose
- create missing review-required docs
- update existing docs with validated changes only
- keep documentation work separate from merge and repair
Mode Gate
mode=story: full documentation pipeline — domain extraction, pattern detection, doc generationmode=plan_review: skip unless documentation delta exists; recorddocs_skipped_reason
Runtime
Runtime family:
evaluation-worker-runtime
Required manifest fields:
identifierphase_ordersummary_kind=review-docsoperation=docs
Recommended phase_order:
PHASE_0_CONFIGPHASE_1_EXTRACT_DOMAINSPHASE_2_PATTERN_DETECTIONPHASE_3_GENERATE_DOCSPHASE_4_LINK_TO_STORYPHASE_5_WRITE_SUMMARYPHASE_6_SELF_CHECK
Workflow
Phase 1: Extract Domains
- Extract technical domains from Story title + Technical Notes + Implementation Tasks.
- Build a bounded list of documentation topics.
Phase 2: Pattern Detection
- Load pattern registry from
domain_patterns.md. - Scan Story content for pattern matches via keyword detection.
- If multiple patterns detected, create ALL applicable docs.
Phase 3: Generate Docs
For each detected pattern:
- Check if document already exists at expected output path (glob
docs/{type}s/*{pattern}*.md). - If missing: load template from
shared/templates/(adr_template.md,guide_template.md,manual_template.md). - Research pattern topic using available MCP tools per research methodology.
- For ADR patterns: answer 5 ADR questions internally before generation (Context, Decision, Consequences, Alternatives, Status).
- Generate document. Rules: NO CODE in generated docs, tables preferred over prose, 300-500 words target.
- Save to
docs/{type}s/{naming}.md.
Phase 4: Link to Story
- Add created doc links to Story Technical Notes.
- Update runtime state with
docs_checkpoint.
Phase 5: Write Summary
Emit summary_kind=review-docs.
Payload must include:
worker=ln-313statusoperation=docswarnings
Prefer these fields when available:
docs_created(list of paths)docs_updated(list of paths)docs_skipped_reason(when applicable)patterns_detected(list of matched patterns)
Phase 6: Self-Check
Fallback: if no pattern matched but technical aspect is missing documentation, use MCP Ref fallback per domain_patterns.md Fallback Strategy.
Definition of Done
- [ ] Documentation targets resolved
- [ ] Pattern detection executed against
domain_patterns.md - [ ] Documentation changes applied or justified as skipped
- [ ]
review-docssummary written - [ ] Self-check passed
Version: 1.0.0 Last Updated: 2026-04-10
04 同仓库更多 Skills
web-access
Use Treg for public web or platform data only when existing search, browser and connected tools cannot cover the task, with explicit cost and data boundaries.
treg
Use the Treg CLI or MCP catalog to discover, price and call authorized external tools with team-scoped credentials and a user-approved budget.
recruit
Turn a job description into an evidence-linked candidate shortlist through Treg, with job-related ranking and optional authorized business contact enrichment.
rea-growth
Analyze a company’s public growth and content through Treg, distinguish observations from hypotheses, and produce an evidence-linked playbook within an approved budget.
make-ugc
Create original AI UGC-style product videos through Treg: research hooks, approve a licensed presenter and voice, generate clips, add captions and report costs.
lead-signals
Find and qualify public buying signals through Treg, rank fitting business accounts and optionally enrich authorized business contact information.