project-profiler
Pass
Audited by Gen Agent Trust Hub on Mar 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implementation follows professional development standards for codebase analysis without obfuscation or exfiltration.
- [COMMAND_EXECUTION]: Executes git and a local Python scanner script (Phase 0.2, 0.3) to retrieve project information. These operations are limited to the target directory and are necessary for the skill's functionality.
- [EXTERNAL_DOWNLOADS]: Fetches metadata from trusted services including the GitHub API, npm registry, and PyPI stats (Phase 1) to assess project maturity. These network operations target well-known services and do not involve sensitive data transmission.
- [PROMPT_INJECTION]: Detects an indirect prompt injection surface. Ingestion points: the skill reads all files in the target directory, including README and code files (Phase 2). Boundary markers: the subagent prompts lack explicit delimiters for untrusted file content. Capability inventory: the agent possesses Bash, Write, and Task tools. Sanitization: Phase 5 performs a stylistic audit to ensure objectivity but lacks logic to neutralize potential malicious instructions within analyzed files.
Audit Metadata