actions-debugging
Pass
Audited by Gen Agent Trust Hub on Jul 20, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection by processing external, untrusted data from GitHub Action logs.
- Ingestion points: Data enters the agent's context via the output of the
github-mcp-server-get_job_logstool as described inSKILL.md. - Boundary markers: The skill does not implement boundary markers or instructions to treat log content as data rather than instructions.
- Capability inventory: The agent is authorized to list and read GitHub Actions and job logs, and is further encouraged to perform local code modifications (e.g., 'Makes the fix in your local codebase') based on the interpreted log content.
- Sanitization: There is no mechanism described for sanitizing or validating the contents of the logs before they are analyzed by the AI.
Audit Metadata