ps-blast-radius
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of code diffs, PR descriptions, and library source code. There are no specific instructions to the agent to treat text within these files strictly as data or to ignore embedded commands.
- Ingestion points: Diff content, PR details, and source files retrieved during dependency tracing (SKILL.md).
- Boundary markers: The instructions do not define delimiters or specific warnings to ignore instructions found within analyzed code.
- Capability inventory: The skill explicitly commands the agent to write and execute scripts on the system to "prove the one fact it's safe because of."
- Sanitization: No sanitization or validation of the input code is specified before execution.
- [DYNAMIC_EXECUTION]: The skill encourages the agent to generate and execute scripts locally to verify facts about the code's safety.
- Evidence: "Write a script or test that runs the real code, run it, and paste what happened" (SKILL.md).
- Context: This represents a standard workflow for coding agents; however, it facilitates the execution of agent-generated code which may carry risks if the agent misinterprets the code's context or if the code being tested has unexpected side effects.
Audit Metadata