Agent skill
doc-adr-audit
Unified ADR audit wrapper that runs validator then reviewer and produces a combined report for fixer consumption
Install this agent skill to your Project
npx add-skill https://github.com/vladm3105/aidoc-flow-framework/tree/main/.claude/skills/doc-adr-audit
Metadata
Additional technical details for this skill
- tags
-
sdd-workflow quality-assurance adr-audit layer-5-artifact shared-architecture
- custom fields
-
{ "layer": 5, "version": "1.0", "priority": "primary", "last_updated": "2026-02-27", "artifact_type": "ADR", "skill_category": "quality-assurance", "development_status": "active", "upstream_artifacts": [ "ADR" ], "downstream_artifacts": [ "Audit Report", "Fix Cycle" ], "architecture_approaches": [ "ai-agent-based" ] } - versioning policy
- tracks ADR-MVP-TEMPLATE schema_version
SKILL.md
doc-adr-audit
Purpose
Run a single ADR audit workflow that executes:
doc-adr-validator(structural/schema gate)doc-adr-reviewer(semantic/content quality gate)
Then emit one combined report optimized for doc-adr-fixer input.
Layer: 5 (ADR Quality Gate Wrapper)
Upstream: ADR file(s)
Downstream:
- Combined Audit Report:
ADR-NN.A_audit_report_vNNN.md - Optional Fix Cycle trigger for
doc-adr-fixer
Why This Skill Exists
Use this wrapper to avoid user confusion between validator and reviewer while preserving separation of concerns.
| Concern | Owner Skill |
|---|---|
| Schema/template compliance | doc-adr-validator |
| Content quality and decision completeness | doc-adr-reviewer |
| Single user-facing audit command | doc-adr-audit |
When to Use
Use doc-adr-audit when:
- You want one command for ADR quality checks
- You need a combined report for
doc-adr-fixer - You are running QA before SYS generation
Do NOT use when:
- ADR does not exist (use
doc-adr/doc-adr-autopilotgeneration first) - You only need one specific check domain (use validator or reviewer directly)
Execution Contract
Input
- ADR path (
docs/05_ADR/ADR-NN_*/...) - Optional: threshold (default review threshold: 90)
Sequence (Mandatory)
1) Run doc-adr-validator
2) Run doc-adr-reviewer
3) Normalize and merge findings
4) Write ADR-NN.A_audit_report_vNNN.md
5) If auto-fixable findings exist, hand off to doc-adr-fixer
Combined Status Rules
PASS: Validator PASS AND Reviewer score >= threshold AND no blocking issuesFAIL: Validator FAIL OR Reviewer score < threshold OR blocking/manual-required issues present
Combined Report Format (for doc-adr-fixer)
Output file: ADR-NN.A_audit_report_vNNN.md
Required sections:
## Summary- ADR ID, timestamp (EST), overall status
- Validator status, reviewer score
## Score Calculation (Deduction-Based)- Formula:
100 - total_deductions - Threshold comparison (
>=90pass gate)
- Formula:
## Validator Findings- List by severity/code
## Reviewer Findings- List by severity/code
## Coverage Findings- Decision completeness summary
- Architecture flow/diagram coverage summary
- Traceability/tag coverage summary
## Fix Queue for doc-adr-fixerauto_fixablemanual_requiredblocked
## Recommended Next Steprun doc-adr-fixer- or
manual update required
Fix Queue Normalization
Each finding MUST include:
source:validator|reviewercode: issue codeseverity:error|warning|infofile: relative pathsection: heading/anchor if knownaction_hint: short imperative guidanceconfidence:high|medium|manual-required
Hand-off Contract to doc-adr-fixer
doc-adr-fixer MUST accept combined audit report as equivalent upstream input:
ADR-NN.A_audit_report_vNNN.md(preferred)ADR-NN.R_review_report_vNNN.md(legacy compatibility)
Precedence rule:
- Select newest timestamp.
- If timestamps are equal, prefer
.A_audit_reportover.R_review_report.
Example Invocation
/doc-adr-audit docs/05_ADR/ADR-01_f1_iam/
Expected outcome:
- validator runs
- reviewer runs
- combined audit report generated
- fixer can execute directly from combined report
Version History
| Version | Date | Changes |
|---|---|---|
| 1.0 | 2026-02-27 | Initial ADR audit wrapper; validator→reviewer orchestration; combined report contract for fixer with .A_ preferred and .R_ legacy compatibility |
Implementation Plan Consistency (IPLAN-004)
- Treat plan-derived outputs as valid source mode and verify intent preservation from implementation plan scope/objectives.
- Validate upstream autopilot precedence assumption:
--iplan > --ref > --prompt. - Flag objective/scope conflicts between plan context and artifact output as blocking issues requiring clarification.
- Do not introduce legacy fallback paths such as
docs-v2.0/00_REF.
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
doc-spec-reviewer
Comprehensive content review and quality assurance for SPEC documents - validates YAML structure, REQ coverage, interface definitions, and identifies issues requiring manual attention
doc-prd-fixer
Automated fix skill that reads review reports and applies fixes to PRD documents - handles broken links, element IDs, missing files, and iterative improvement
doc-procspec-autopilot
Automated PROCSPEC (Process Specification) generation from REQ - generates specifications for SOPs, runbooks, playbooks, and operational procedures
doc-riskspec-autopilot
Automated RISKSPEC (Risk Specification) generation from REQ - generates specifications for risk matrices, impact assessments, and mitigation plans
doc-ptest-autopilot
Automated PTEST generation and review orchestration for performance category and threshold validation
doc-adr-autopilot
Automated ADR generation pipeline from BRD Architecture Decision Requirements - analyzes topics, generates Context-Decision-Consequences records
Didn't find tool you were looking for?