Run a documentation-focused maintenance checkup of the Chatto codebase. Fans out to /fdr, /adr, and /chatto-architecture-inventory, then compiles a single consolidated report. Always propose-only; no changes are applied without explicit user approval. At the end, points the human at other maintenance skills (/update-project-dependencies, /chatto-security-review) they may want to run themselves.
69
83%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
View guide
Passed
No findings from the security scan
The standard documentation rounds. The checkup is propose-only: it audits and reports, never auto-applies fixes.
Heavier or more opinionated work — dependency upgrades, security review — is deliberately not invoked here. The end-of-report section points the human at those skills so they can decide whether to run them.
Run these in parallel via the Skill tool:
/fdr (no args) — audit all FDRs against the codebase. Surface discrepancies, stale design decisions, and any user-facing features that should now have an FDR./adr — audit ADRs for staleness. Surface cited file paths or APIs that no longer exist; flag superseded ADRs still referenced as authoritative./chatto-architecture-inventory — audit docs/architecture/ against the current state of runtime components, projections and snapshots, EVT event tokens, EVT/live subject patterns, NATS resources and runtime-state keys, durable effects, mounted interface boundaries, and realtime delivery. Report drift; propose updates.docs/architecture/, grouped by inventory category./update-project-dependencies — bump deps within semver and run tests/chatto-security-review — multi-agent security auditOutput a single Markdown report. Brief findings, no walls of text. For every finding, include:
/update-project-dependencies or /chatto-security-review. They're surfaced as pointers at the end of the report; the human invokes them.65f65c3
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.