pma
Pass
Audited by Gen Agent Trust Hub on Apr 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill defines a rigorous development lifecycle with built-in quality controls, such as a mandatory investigation phase and explicit approval gates before implementation.\n- [COMMAND_EXECUTION]: The skill provides guidelines for using bash and tmux to manage local development processes and servers. These commands are localized to the current directory and use unique session names based on the directory path's md5 hash.\n- [PROMPT_INJECTION]: The skill involves reading and analyzing project files, identifying a surface for indirect prompt injection. This is mitigated by the workflow's requirement for human approval during the proposal phase.\n
- Ingestion points: Project source code, configuration, tests, and documentation (e.g., docs/task/, docs/plan/, docs/changelog.md).\n
- Boundary markers: Absent; the skill does not instruct the agent to use specific delimiters for external content.\n
- Capability inventory: Local command execution (bash, tmux), Git operations, and file system write access.\n
- Sanitization: Absent; the workflow relies on the 'Proposal' phase gate and focused verification.
Audit Metadata