skill-management

Pass

Audited by Gen Agent Trust Hub on Feb 24, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill recommends using pnpm dlx skills add oakoss/agent-skills for skill installation and discovery. This command interacts with the official 'skills' CLI and the author's repository, which is a standard procedure for skill management within this ecosystem.\n- [PROMPT_INJECTION]: The skill outlines a systematic procedure for auditing external content (other agent skills), which creates an indirect prompt injection surface as the agent is instructed to ingest and process files that could contain malicious instructions.\n
  • Ingestion points: The auditing process defined in references/auditing-skills.md (Phase 1) requires reading all files in a target skill directory, including SKILL.md, references, and scripts.\n
  • Boundary markers: The instructions do not specify any delimiters or safety markers for the content being analyzed during the audit phases.\n
  • Capability inventory: The framework allows the agent to conduct systematic reviews and utilize the pnpm validate:skills command on audited skills.\n
  • Sanitization: The skill relies on the 9-phase audit process itself as a validation mechanism rather than implementing automated sanitization of the input data.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 24, 2026, 08:36 PM