using-aegis — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited using-aegis (Agent Skill) and scored it 45/100 (orange). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 2 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 2 flagged
A base64 string of 128+ characters appears in a documentation file. Encoded prompt injection hides the hostile instruction in base64 — invisible to keyword filters — and relies on the agent's ability to decode it at runtime. There is no normal authoring reason to embed a multi-hundred-byte base64 blob in skill docs.
*.sig, SIGNATURES) outside the documentation.A base64 string of 128+ characters appears in a documentation file. Encoded prompt injection hides the hostile instruction in base64 — invisible to keyword filters — and relies on the agent's ability to decode it at runtime. There is no normal authoring reason to embed a multi-hundred-byte base64 blob in skill docs.
*.sig, SIGNATURES) outside the documentation.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.
<SUBAGENT-STOP>Subagents skip this skill.</SUBAGENT-STOP>
<EXTREMELY-IMPORTANT> You have Aegis.
Before any response/action, check if an Aegis skill is explicit or clearly relevant. Load only that skill; otherwise proceed normally. </EXTREMELY-IMPORTANT>
(README/ADR/rules/docs/aegis/baseline). If none fit, bounded index-first scan; create a baseline only with evidence, and still answer.
/aegis-goal or Aegis goal: loads goal-framing for goal, successevidence, stop condition, and non-goals before onward routing.
+ baseline check + TDD Route + verification. Medium/high: baseline read-set + plan. TDD Route: auto=strict/light/skipped; off=no automatic TDD, verification stays. Add Spec Brief or Design Spec only when complexity, ambiguity, contracts, or cross-module impact require it. Contract/shared/core or cross-module changes are never low without evidence.
ArchitectureReviewRequired: yes for medium/high, architecture,contract, cross-module, owner, source-of-truth, fallback/adapter, or project-baseline tasks. Carry it to verification-before-completion.
edits never write project files. Baseline/spec/plan/work records use configured Aegis workspace support only when persistent evidence is needed.
candidates, not prompt payloads: summarize first; for large inputs use bounded index→window→excerpt.
history.jsonl,.codex/sessions, ~/.claude/projects, or large logs by default. Only read direct evidence when requested or required, with scope/time/line bounds.
Contract when useful: Route: fast-path; Aegis Reason Note; Why; Next.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.