- Home
- Skills
- Vladm3105
- Aidoc Flow Framework
- Doc Ears Reviewer
doc-ears-reviewer_skill
- Python
9
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill vladm3105/aidoc-flow-framework --skill doc-ears-reviewer- SKILL.md15.3 KB
Overview
This skill performs comprehensive content review and quality assurance for EARS (Easy Approach to Requirements Syntax) documents. It validates folder structure, requirement syntax, threshold quantification, PRD traceability, and testability. The reviewer outputs a scored report, an issue list, and updates a mandatory upstream drift cache.
How this skill works
The reviewer loads EARS files (single or sectioned), enforces nested-folder structure, and runs layered checks: EARS syntax, thresholds, PRD alignment, testability, placeholders, section completeness, naming, and upstream drift detection. Auto-fixable structural issues are applied automatically; content issues are flagged for manual remediation. A review score is calculated and a versioned report and drift cache are written alongside the EARS file.
When to use it
- Immediately after EARS generation (post-autopilot)
- After manual edits to EARS documents
- Before running BDD generation or doc-bdd-autopilot
- As a periodic QA gate in CI/CD documentation pipelines
- When verifying PRD traceability and upstream freshness
Best practices
- Run only after structure validation (nested folders required)
- Keep PRD references (@prd/@ref) accurate to enable drift detection
- Quantify all timing/performance thresholds (WITHIN clauses with units)
- Mark deferred items explicitly; avoid placeholders like [TODO]/[TBD]
- Persist and review the drift cache with every run to detect upstream changes
Example use cases
- Final QA pass before BDD generation to ensure requirements are testable and traceable
- Automated review step in a docs CI pipeline to block regressions from upstream PRD edits
- Reviewing a manually edited EARS file to catch ambiguous subjects and missing thresholds
- Generating a versioned review report and updated drift cache after a review
- Flagging orphaned EARS requirements that lack PRD mapping for manual triage
FAQ
Structure compliance is blocking: the reviewer stops other checks and reports errors until the nested-folder rule is satisfied.
Is the drift cache optional?
No. The upstream drift cache is mandatory and must be created or updated on every review to track upstream PRD changes and detect stale EARS content.