The Agent Skills Directory

[SAFE]: The skill follows security best practices by explicitly instructing the agent to avoid committing or logging credentials (Stage 7 and Invariants in SKILL.md). It also enforces 'Gate 4' in its protocol which explicitly checks for security advisories and pinned dependencies to maintain supply chain hygiene.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external registries and changelogs during verification stages (Stage 2 and Stage 8 in SKILL.md). While it lacks explicit prompt boundary markers or sanitization for this external data, the risk is minimized by the skill's strictly procedural validation logic and the requirement for user authorization before any irreversible publishing actions are taken (Stage 7).

fable-release