setup — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited setup (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
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.
A guided interview that turns the blueprint into a personalized AIOS. Read aios-intake.md first; if it's filled, confirm and scaffold; if not, run the interview conversationally, one question at a time, with a recommended answer so the user can confirm/correct.
aios-intake.md as you go)context/about-me.md, CLAUDE.md header)references/voice.md)context/priorities.md)context/about-business.md, Knowledge base)connections.md)context/about-team.md)CLAUDE.md Guardrails)Fill every {{placeholder}} in CLAUDE.md (name, persona, knowledge base, voice summary, connections, guardrails — customize rule #1 and #8 to the domain). Write context/*, references/voice.md, and the connections.md rows. Back up any existing versions to archives/ first.
Point the user at: /structure (frame fuzzy asks) · /triage · /daily · /weekly (Cadence) · /wiki (second brain: raw/ → wiki/) · /graph · /graph-query · /graph-ingest (knowledge graph) · /workflow (dynamic workflows) · /experiment (autoresearch loop) · /audit + /level-up (weekly). Then suggest the first prompt: "what should I focus on this week?"
Offer to set up the knowledge graph so the user can see and query their code + second brain: run python tools/graphify_setup.py install (installs graphify and wires the Claude Code & Codex skills). Then /graph builds it. Skippable and additive — graphs code + the committed `wiki/` only, never raw/.
aios-intake.md and re-running.~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.