agency-minimal-change-engineer
Pass
Audited by Gen Agent Trust Hub on Aug 11, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: No security issues were identified in the skill. The content consists entirely of markdown-based identity and workflow instructions for a coding assistant.
- [PROMPT_INJECTION]: The skill defines an interaction model where the agent processes external task descriptions and codebase content, which constitutes a typical data ingestion surface.
- Ingestion points: The agent reads project source code and user-provided task descriptions from the local environment.
- Boundary markers: The instructions establish clear boundaries by explicitly commanding the agent to 'touch only what the task requires' and 'ignore' unrelated code or improvements.
- Capability inventory: The skill is designed for agents with file-system read and write capabilities, as documented in the workflow requirements.
- Sanitization: The skill relies on strict instructional constraints and a 'minimal surface area' philosophy to mitigate risks from processing potentially untrusted data.
Audit Metadata