elite-ops
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEPROMPT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill uses directive language to alter the agent's core behavioral model. Evidence: The instruction 'This overrides passive assistant behavior with owner-engineer execution posture' in SKILL.md explicitly commands the agent to bypass its default operational state.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and act upon data from external code repositories, which serves as a potential vector for malicious instructions.
- Ingestion points: The 'Repo Findings' section (D) in SKILL.md directs the agent to 'Assimilate the codebase before touching it.'
- Boundary markers: Absent. No instructions are provided to treat repository content as data rather than instructions.
- Capability inventory: The agent is empowered to 'Execute the actual changes' (Section F), which includes writing or modifying files.
- Sanitization: Absent. There are no guidelines for validating or sanitizing the content retrieved from the repository before implementation.
Audit Metadata