receive-feedback
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- Prompt Injection (LOW): The skill is designed to ingest and process code review feedback from external sources, creating a surface for indirect prompt injection.
- Ingestion points: External feedback items provided by reviewers (referenced in SKILL.md and VERIFICATION.md).
- Boundary markers: The instructions do not define specific delimiters or isolation techniques (like XML tags or 'ignore' warnings) to prevent feedback from being interpreted as instructions.
- Capability inventory: The skill uses file reading (
Read), searching (Grep), and skill loading (Skill). It also recommends running tests or scripts for verification. - Sanitization: No explicit sanitization or filtering is prescribed; the skill relies on the agent's technical verification logic.
- Dynamic Execution (LOW): VERIFICATION.md instructs the agent to 'reproduce with test or script' when validating bugs. Running code generated or influenced by untrusted external feedback is a sensitive operation, though standard for code review tasks.
Audit Metadata