preflight-checks
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION] (LOW): Indirect prompt injection vulnerability surface detected. The skill instructions suggest checking
README.mdfor documented security commands and CI workflows for flags to replicate locally. If these files are attacker-controlled, they could influence the agent's behavior. - Ingestion points:
README.md,.github/workflows/*.yml - Boundary markers: Absent. No instructions to ignore malicious commands or validate input.
- Capability inventory: High-privilege shell execution via
docker,npx, and language-specific package managers. - Sanitization: None. Commands are executed based on findings in the local environment.
- [EXTERNAL_DOWNLOADS] (SAFE): The skill relies on external tool runners like
docker(pullingsemgrep/semgrep) andnpx(pulling various JS tools). These are appropriate for the skill's primary function of code validation and quality checking. - [COMMAND_EXECUTION] (SAFE): The skill facilitates the execution of local development tools (
pytest,go test,cargo clippy). These commands are standard for the described use case.
Audit Metadata