lean-instructions

Pass

Audited by Gen Agent Trust Hub on Nov 10, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted repository data to derive and update project instructions. Malicious instructions hidden in repository manifests, scripts, or documentation could influence the generated content or the verification steps.
  • Ingestion points: SKILL.md (Step 1) reads existing AGENTS.md, CLAUDE.md, manifests, scripts, CI configs, and general documentation.
  • Boundary markers: Absent. There are no explicit instructions to the agent to treat content within ingested files as data only and to ignore embedded instructions.
  • Capability inventory: The skill is capable of reading/writing files at the repository root and executing shell commands to verify project-specific instructions.
  • Sanitization: The skill includes logic to inspect scripts before execution and explicitly warns against including secrets or guesses in the output.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 10, 2026, 02:14 PM