← ClaudeAtlas

documentation-sync-enginelisted

Canonical orchestrator that runs the fast stale-doc detector and dispatches to per-domain doc-sync skills for the affected canonical paths.
Eliyce/paqad-ai · ★ 4 · AI & Automation · score 76
Install: claude install-skill Eliyce/paqad-ai
## What It Does Acts as the canonical entry point for post-implementation documentation sync. Runs `diff-doc-sync` first to narrow to actually-stale canonical doc paths, then dispatches each path to the domain-specific maintainer (api, integration, error catalog, canonical, or glossary). Aggregates the results into a single report so the caller does not have to invoke five skills sequentially. The point is to give callers one place to ask "sync the docs after this change" and get consistent, deduplicated output every time. ## Use This When Use this after implementation lands (or just before handoff) on `feature-development`, `bug-fix`, `refactor`, or `migration` workflows. Skip when `differential_refresh` is disabled in the project profile and the team prefers running the per-domain skills directly. ## Inputs - Read the changed file list at `changed_files`. - Use the detector at `detector_script_path` (via `diff-doc-sync`) to produce the stale canonical doc set. - Read `references/orchestration-rules.md` before dispatching so the routing rules and execution order stay consistent across runs. - When `target_domains` is non-empty, intersect the dispatched set with the requested domains. ## Procedure 1. Invoke `diff-doc-sync` with the changed file list to produce the candidate stale doc set (JSON array). 2. Pipe that array's paths into `scripts/route-paths.sh` to assign each path a delegate domain (`api | integration | error | glossary | canonical`); see `assets/routing-