regression-testing
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data such as bug descriptions and pull request details to generate or select tests. Maliciously crafted input in these fields could potentially influence the agent's decision-making process.
- Ingestion points: The 'bug' object description in the 'Bug Regression Test' task and the 'changedFiles' list in the 'selectTests' function.
- Boundary markers: There are no explicit delimiters or instructions to isolate external text from system instructions in the provided snippets.
- Capability inventory: Task execution and agent coordination via the 'FleetManager' allow the agent to perform automated actions based on input data.
- Sanitization: No explicit sanitization or validation logic is defined for the content of external inputs.
- [EXTERNAL_DOWNLOADS]: The skill references external components for CI/CD workflows and data validation.
- Documentation examples reference the 'dorny/paths-filter' GitHub Action.
- The JSON schema points to a remote resource at 'https://agentic-qe.dev/'. These references are informative and consistent with the skill's primary testing functionality.
- [COMMAND_EXECUTION]: The documentation provides examples of running test suites in CI environments.
- Examples include 'npm run test:affected', 'npm run test:smoke', and 'npm test -- --coverage'. These are standard development commands but involve shell execution in the targeted environment.
Audit Metadata