build-scenario-tests
Pass
Audited by Gen Agent Trust Hub on Aug 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill explicitly instructs the agent to execute shell commands to run test suites, linters, and build processes (as seen in Workflow Step 5: 'Run the new test alone... Then run the nearest relevant suite and the repository's required lint, type, build, and test commands'). This is a core part of verifying the generated tests.
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests untrusted data from the local repository and has the capability to write code and execute commands based on that data.
- Ingestion points: The agent is instructed to read local files such as README.md, AGENTS.md, and contribution guides (Workflow Step 1).
- Boundary markers: The skill does not provide instructions to use delimiters or ignore instructions embedded within the ingested files.
- Capability inventory: The agent can write new test files to the local file system and execute shell commands to run runners and build tools.
- Sanitization: No explicit sanitization or validation of the ingested content is mentioned.
Audit Metadata