doc-ctr-fixer_skill

This skill automatically reads the latest review report and applies fixes to CTR documents, aligning structure, links, IDs, and content.

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-ctr-fixer

SKILL.md26.1 KB

Overview

This skill automates fixes to CTR (Contract) documents by reading the latest review report and applying targeted corrections. It handles folder structure, missing companion files, broken links, element ID conversions, content placeholders, dual-file synchronization (MD ↔ YAML), and iterative upstream drift merging. The goal is to move CTRs from review findings to a validated, versioned state ready for re-review or release.

How this skill works

The skill locates the most recent review report for a given CTR and parses categorized issues. It runs a series of ordered fix phases: enforce nested folder structure, create missing files from templates, repair links and YAML refs, normalize element IDs, fill basic content gaps, sync MD and YAML companions, and perform tiered auto-merge for upstream REQ drift. Each change is recorded in a fix report and the skill can invoke a re-review loop until a quality threshold is reached.

When to use it

After running the reviewer to auto-apply identified fixes
As part of CI/CD quality gates to ensure CTR consistency before merge
When MD and YAML companion files are out of sync or missing
To batch-fix multiple CTRs based on generated review reports
When upstream REQ changes require controlled auto-merge and version updates

Best practices

Run doc-ctr-reviewer first to produce the review report; fixer requires that input
Keep a drift cache for each CTR so upstream changes can be measured and traced
Use the skill in iterative cycles: Fix → Re-run reviewer → Repeat until quality threshold met
Limit automatic edits to syntactic fixes; flag placeholders and ambiguous content for manual completion
Enable backups or source control commits for each fix phase to allow rollbacks

Example use cases

Fix CTR moved into an incorrect folder: create nested folder, move MD/YAML, update relative links
Auto-generate missing YAML companion from MD using templates and link the files
Repair broken internal schema refs and convert absolute paths to correct relative $ref entries
Convert legacy element IDs (API-XXX, EVT-XXX) to valid CTR type codes and update references
Detect upstream REQ drift under 5% and auto-merge changes with a patch version bump

FAQ

No. The skill auto-fixes structural, syntactic, and reference issues. Placeholders like [TODO] or ambiguous content are flagged for manual authoring and not auto-completed.

How does upstream drift get handled?

Drift is measured by comparing stored snapshots to current REQ content. Small changes (<5%) are auto-merged with a patch bump; moderate changes (5–15%) include changelogs and minor bumps; major drift (>15%) triggers archive and regeneration workflows.