flaky-test-catcher
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external content from GitHub Actions CI logs and repository commit messages.
- Ingestion points: Reads log data using
gh run view --logand commit metadata viagh api. - Boundary markers: The skill instructs the agent to use
grepto isolate specific failure patterns (--- FAIL:) and limited context, which constrains the volume of ingested data. However, there are no explicit instructions for the agent to ignore potential instructions embedded within these external sources. - Capability inventory: The skill can perform GitHub API operations including listing jobs, fetching commits, and creating issues. It does not have general file system write access or unrestricted network capabilities.
- Sanitization: The skill lacks explicit sanitization or escaping protocols for the log excerpts and commit messages before they are interpolated into the GitHub issue body.
- [COMMAND_EXECUTION]: The skill executes shell commands using the
gh(GitHub CLI),git, andgreputilities. These are used for their intended administrative and diagnostic purposes (e.g., retrieving repository metadata, fetching CI logs, and filtering text) and do not involve executing untrusted scripts.
Audit Metadata