visual-review

Fail

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill instructs the agent to run 'bunx impeccable detect', which downloads and executes code from the npm registry at runtime. The package is not from a trusted organization or well-known service.
  • [COMMAND_EXECUTION]: The skill references and executes 'scripts/skills-browser.sh', which is not contained within the skill's provided files.
  • [EXTERNAL_DOWNLOADS]: The skill uses 'bunx' to fetch external dependencies from public registries.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from the user's repository, including source code and terminal output. 1. Ingestion points: Changed files and CLI output. 2. Boundary markers: Absent. 3. Capability inventory: Shell execution and network access. 4. Sanitization: None identified.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Sep 11, 2026, 02:01 AM
Security Audit — agent-trust-hub — visual-review