sync — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited sync (Agent Skill) and scored it 96/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 1 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 1 flagged
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
This skill updates the AI-tooling layer of an existing Trabuco project so its skills, subagents, prompts, hooks, and review scaffolding match what the currently-installed Trabuco CLI would generate today.
mcp__trabuco__get_project_info to confirm it's a Trabuco project — if it returns an error, tell the user and stop.mcp__trabuco__sync_project with apply: false to produce a plan. The response is JSON with:would_add: files the current CLI would emit that are missingalready_present: files that exist (untouched regardless of content)out_of_jurisdiction: business / infra files sync never considersblockers: reasons sync can't proceed (e.g., no .trabuco.json)would_add list grouped by directory (.claude/, .codex/, .cursor/, .github/, .ai/, top-level). Keep the explanation short — the plan output speaks for itself.would_add is non-empty, ask the user "Apply these N additions?" Wait for explicit confirmation (not just acknowledgement — "yes" or "apply", not "sounds good").mcp__trabuco__sync_project with apply: true. Report the count of files added.trabuco doctor to confirm project health."git diff before committing.".trabuco.json).CLAUDE.md), explain that sync is additive-only and the workaround is to delete the file first, then run sync.application.yml, docker-compose.yml, migrations, and CI workflows (other than copilot-setup-steps.yml) are never considered. If the user asks sync to "update my application.yml," redirect them.User: my project feels stale, can you update the AI helpers? You: [call get_project_info, then sync_project with apply=false] You: Sync would add 4 files to /path/to/project: -.claude/skills/add-test/SKILL.md-.claude/subagents/code-simplifier.md-.ai/prompts/testing-guide.md-.cursor/rules/add-endpoint.mdcApply these? User: yes You: [call sync_project with apply=true] You: Added 4 files. Restart your Claude Code session to pick up the new skills and subagents.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.