gabe-push — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited gabe-push (Agent Skill) and scored it 45/100 (orange). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 1 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 1 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.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.
Expose /gabe-push as a selectable skill in agents that use skills instead of native slash-command routing. The command markdown remains the source of truth; this wrapper only tells the host how to find and execute it.
$ARGUMENTS..claude/commands/gabe-push.md from the current project, if present.agents/commands/gabe-push.md from the current project, if present~/.claude/commands/gabe-push.md~/.agents/commands/gabe-push.md~/projects/gabe_lens/commands/gabe-push.mdhandling, push/PR/CI/promotion behavior, deployment logging, PLAN auto-tick, and any visible output requirements.
native slash command, load that command's spec from the same search order and follow its output contract exactly.
spec.
When both Claude and Codex installed assets exist, prefer the path for the active host. Claude Code should use ~/.claude; Codex should use ~/.agents; repo-local .claude/.agents mirrors outrank home installs when present.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.