senior-implementer
Pass
Audited by Gen Agent Trust Hub on May 20, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious behavior or technical vulnerabilities were detected in the skill instructions.
- [PROMPT_INJECTION]: No direct injection attempts were found. The skill possesses an indirect prompt injection surface as it is designed to process untrusted external briefs and project convention files. Ingestion points: external briefs and convention files such as CLAUDE.md or .cursorrules (SKILL.md). Boundary markers: absent. Capability inventory: file-read, file-write, and command execution for feedback loops (SKILL.md). Sanitization: absent.
- [COMMAND_EXECUTION]: The skill encourages running project-specific feedback loops like tests and linters, which is appropriate for its stated purpose as a coding assistant.
- [DATA_EXFILTRATION]: No patterns for accessing sensitive system files or transmitting data to external domains were identified.
Audit Metadata