audit-skill-lifecycle

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) through its deep semantic analysis of external plugin files.\n
  • Ingestion points: Skill directories, agent files, command files, reference documents, and data files (JSON, YAML, Markdown) parsed during the Discovery phase in SKILL.md.\n
  • Boundary markers: Absent; the skill workflow does not specify the use of delimiters or 'ignore' instructions to isolate the content of audited files from the agent's logic.\n
  • Capability inventory: The skill performs extensive file system reads across the plugin structure and writes audit artifacts, reports, and recommendations to the '.claude/audits/' directory.\n
  • Sanitization: Absent; there is no evidence of instruction filtering, validation, or escaping of the parsed content before it is analyzed by the model.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 08:41 AM