idk-now

Pass

Audited by Gen Agent Trust Hub on Oct 23, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill analyzes untrusted repository files to determine project vision and suggest next steps, creating a surface where embedded instructions could potentially influence agent behavior.\n
  • Ingestion points: Reads various repository files including README.md, PRODUCT.md, VISION.md, AGENTS.md, and CLAUDE.md, as well as git logs.\n
  • Boundary markers: The instructions do not define delimiters or specific 'ignore' markers for the content read from files.\n
  • Capability inventory: The skill utilizes git log and find for surveying and has the ability to recommend and initiate other skills.\n
  • Sanitization: No specific validation or sanitization steps are documented for the data retrieved from the file system.\n- [COMMAND_EXECUTION]: The skill performs shell operations to gather environment data and inventory available tools.\n
  • Evidence: Executes git log --oneline -15 to survey the repository history.\n
  • Evidence: Uses find to identify SKILL.md files across the project workspace.\n
  • Evidence: Utilizes npx skills list to check for installed skill packages.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 23, 2026, 01:31 AM