lov-code-review

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external, untrusted code differences (diffs), creating a surface for indirect prompt injection attacks where malicious code could contain instructions aimed at the reviewing agent. \n
  • Ingestion points: SKILL.md defines a workflow that begins by reading diffs and project rules from the local environment. \n
  • Boundary markers: There are no explicit instructions or delimiters defined to isolate the untrusted diff content or to instruct the agent to ignore any embedded directives within that content. \n
  • Capability inventory: The skill leverages scripts/profile_store.py for writing to the filesystem and explicitly mentions utilizing the host agent's CLI and browser capabilities. \n
  • Sanitization: No sanitization or validation logic is specified for the input data before it is processed by the agent.\n- [EXTERNAL_DOWNLOADS]: The documentation provides installation instructions using npx to add the lov-code-review package from the NPM registry.\n
  • Evidence: README.md contains the command 'npx skills add lov-code-review -g -y'.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 09:50 AM
Security Audit — agent-trust-hub — lov-code-review