gentle-ai-rdd-defect-workflow
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external, potentially untrusted data from repository issues and pull requests.
- Ingestion points: The workflow instructions in
SKILL.md(Execution Steps 1 and 2) direct the agent to analyze issue approvals and pull request conflicts from the repository. - Boundary markers: There are no instructions to use delimiters or specific markers to isolate or ignore instructions embedded within the external issue or pull request content.
- Capability inventory: The skill (Execution Step 4) requires the agent to execute commands, run tests, and record results. This provides a mechanism where instructions embedded in untrusted external data could potentially trigger unintended actions during test execution or CLI interaction.
- Sanitization: No sanitization, escaping, or validation steps for the external content are specified before it is interpolated into the agent's workflow.
Audit Metadata