OpenSkillIndex← back to search
claude-code

Claude-code-skills ln-314-review-repair-worker

Use when accepted findings require bounded repair changes and a structured repair summary.

install
source · Clone the upstream repo
git clone https://github.com/levnikolaevich/claude-code-skills
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/levnikolaevich/claude-code-skills "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills-catalog/ln-314-review-repair-worker" ~/.claude/skills/levnikolaevich-claude-code-skills-ln-314-review-repair-worker && rm -rf "$T"
manifest: skills-catalog/ln-314-review-repair-worker/SKILL.md
source content

Paths: File paths (

shared/
, 
references/
, 
../ln-*
) are relative to skills repo root.

Type: L3 Worker Category: 3XX Planning

Review Repair Worker

Mandatory Read

MANDATORY READ: Load 

shared/references/evaluation_worker_runtime_contract.md
, 
shared/references/evaluation_summary_contract.md
, 
shared/references/cleanup_evidence_contract.md
MANDATORY READ: Load 
../ln-310-multi-agent-validator/references/context_review_pipeline.md
, 
../ln-310-multi-agent-validator/references/penalty_points.md
MANDATORY READ: Load 
../ln-310-multi-agent-validator/references/structural_validation.md
, 
../ln-310-multi-agent-validator/references/standards_validation.md
, 
../ln-310-multi-agent-validator/references/solution_validation.md
, 
../ln-310-multi-agent-validator/references/workflow_validation.md
, 
../ln-310-multi-agent-validator/references/quality_validation.md
, 
../ln-310-multi-agent-validator/references/dependency_validation.md
, 
../ln-310-multi-agent-validator/references/cross_reference_validation.md
, 
../ln-310-multi-agent-validator/references/risk_validation.md
, 
../ln-310-multi-agent-validator/references/premortem_validation.md
, 
../ln-310-multi-agent-validator/references/traceability_validation.md

Purpose

  • apply accepted low-risk repairs using the 11-group systematic validation structure
  • keep repairs separate from aggregation and approval
  • record cleanup evidence if background tools are launched

Mode Gate

  • mode=story
    : full 11-group repair using findings from ln-312
  • mode=plan_review|context
    : apply accepted corrections only (max 5 corrections per 
    ../ln-310-multi-agent-validator/references/context_review_pipeline.md
    Compare & Correct Safety Rules)

11-Group Validation Structure (mode=story)

The checklists below are mandatory inputs loaded above. Use the table only as the routing map for applying them.

#GroupCriteriaChecklist
1Structural#1-#4, #23-#24
../ln-310-multi-agent-validator/references/structural_validation.md
2Standards#5
../ln-310-multi-agent-validator/references/standards_validation.md
3Solution#6, #21, #28
../ln-310-multi-agent-validator/references/solution_validation.md
4Workflow#7-#13
../ln-310-multi-agent-validator/references/workflow_validation.md
5Quality#14-#15
../ln-310-multi-agent-validator/references/quality_validation.md
6Dependencies#18-#19/#19b
../ln-310-multi-agent-validator/references/dependency_validation.md
7Cross-Reference#25-#26
../ln-310-multi-agent-validator/references/cross_reference_validation.md
8Risk#20
../ln-310-multi-agent-validator/references/risk_validation.md
9Pre-mortem#27
../ln-310-multi-agent-validator/references/premortem_validation.md
10Verification#22
../ln-310-multi-agent-validator/references/traceability_validation.md
11Traceability#16-#17, #17b-#17c
../ln-310-multi-agent-validator/references/traceability_validation.md

Repair Rules

  • Zero out penalty points only when the defect is actually repaired (not just acknowledged).
  • Use 
    FLAGGED
    only when human judgment is required and auto-fix cannot safely continue.
  • Maximum penalty per 
    ../ln-310-multi-agent-validator/references/penalty_points.md
    (do not hardcode).
  • Test strategy section may exist but remain empty.
  • Apply auto-fix actions exactly as specified in each checklist's "Auto-fix actions" column.

Runtime

Runtime family:

  • evaluation-worker-runtime

Required manifest fields:

  • identifier
  • phase_order
  • summary_kind=review-repair
  • operation=repair

Recommended 

phase_order
:

  1. PHASE_0_CONFIG
  2. PHASE_1_LOAD_FINDINGS
  3. PHASE_2_GROUP_STRUCTURAL
    (#1-#4, #23-#24)
  4. PHASE_3_GROUP_STANDARDS_SOLUTION
    (#5, #6, #21, #28)
  5. PHASE_4_GROUP_WORKFLOW_QUALITY
    (#7-#15)
  6. PHASE_5_GROUP_DEPS_XREF
    (#18-#19/#19b, #25-#26)
  7. PHASE_6_GROUP_RISK_PREMORTEM
    (#20, #27)
  8. PHASE_7_GROUP_VERIFICATION_TRACE
    (#22, #16-#17, #17b-#17c)
  9. PHASE_8_VERIFY_LOCAL_RESULT
  10. PHASE_9_WRITE_SUMMARY
  11. PHASE_10_SELF_CHECK

Summary

Emit 

summary_kind=review-repair
.

Payload must include:

  • worker=ln-314
  • status
  • operation=repair
  • warnings

Prefer these fields when available:

  • penalty_before
    (from ln-312 findings)
  • penalty_after
    (after repairs)
  • flagged_items
    (list of items requiring human judgment)
  • coverage_summary
    (AC coverage percentage)
  • groups_processed
    (count)

Definition of Done

  • Findings loaded from ln-312
  • All 11 groups processed (mode=story) or accepted corrections applied (mode=plan_review|context)
  • Penalty before/after tracked
  • FLAGGED items recorded
  • Local verification completed
  • Cleanup evidence recorded when needed
  • review-repair
    summary written
  • Self-check passed

Version: 1.0.0 Last Updated: 2026-04-10