blast-radius
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to create and run scripts or tests against the local codebase to prove safety facts and identify potential regressions.
- Evidence: The 'How sure are you' section (Step 4: 'You ran it') and the 'Steps' section (Step 5: 'Write a script or test that runs the real code, run it').
- [DYNAMIC_EXECUTION]: The instructions guide the agent to generate and execute code at runtime to verify logic and reproduce issues.
- Evidence: 'Prove the one fact. Write a script or test that runs the real code, run it, and paste what happened.'
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external data such as pull request diffs, commits, and third-party library source code, which could contain malicious instructions aimed at the agent.
- Ingestion points: The agent ingests PR diffs, commits, and external library source code as specified in Step 1 and Step 3.
- Boundary markers: No explicit boundary markers or 'ignore' instructions are provided for the data being analyzed.
- Capability inventory: The agent has the capability to write files and execute commands to run the verification tests.
- Sanitization: No explicit sanitization or filtering of the ingested code content is mentioned.
Audit Metadata