SaferSkills independently audited idem (Agent Skill) and scored it 92/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 2 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
MEDIUMInstruction telling the agent not to ask for approvalSS-SKILL-INJECT-DONT-ASK-01 · Prompt injection · CHANGELOG.md×2
MEDIUM — it fires on intent; the real damage depends on the host agent's own approval-gating.
Why it matters
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.
The exact value spotted
excerptCHANGELOG.md· markdown
1716- **`/vibe-upgrade` is now full-featured** (67 → 353 lines), a complete upgrade
1717workflow that keeps vibestack's fast-forward-only safety:
1718- Inline upgrade consent gate (Yes / Always / Not now / Never ask again),
1719triggered by an `UPDATE:` line in the preamble.
1720- Auto-upgrade mode via `VIBESTACK_AUTO_UPGRADE` or the `auto_upgrade` config
Occurrences
2 occurrences · first at L1718, also L1724Show all 2 locations
Line
File
L1718
CHANGELOG.md
L1724
CHANGELOG.md
How to fix
Remove the approval-skipping instruction, or scope it narrowly to a specific safe, reversible action.
Delete blanket "don't ask / no need to confirm" directives from the skill.
If the skill is a genuine autonomous job, restrict the opt-out to a named non-destructive action rather than all actions.