legacy-summary
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEDYNAMIC_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DYNAMIC_EXECUTION]: The skill explicitly directs the agent to generate and run characterization scripts at runtime to observe code behavior.
- [COMMAND_EXECUTION]: Executing the generated scripts involves invoking language runtimes or shell commands in the local environment.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted legacy source code as its primary input, creating a potential surface for injection. 1. Ingestion points: Legacy source code files targeted for audit. 2. Boundary markers: No explicit delimiters or instructions are provided to the agent to distinguish between code logic and potential embedded instructions. 3. Capability inventory: Local script generation and shell command execution. 4. Sanitization: The instructions do not define any sanitization or filtering for the audited code.
Audit Metadata