github-issue-agent
Pass
Audited by Gen Agent Trust Hub on Aug 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a security-first design by defaulting all issue creation and modification operations to a dry-run mode, ensuring no live changes occur to the repository without explicit intent.
- [SAFE]: A dedicated secret redaction mechanism (redaction_gate.py) is integrated into the workflow to scan issue content for sensitive tokens, API keys, and private keys before submission.
- [CREDENTIALS_UNSAFE]: The 'tests/test_secret_redaction.py' file contains hardcoded strings mimicking the format of GitHub tokens and RSA private keys. These are non-functional dummy values used exclusively as test vectors for the redaction engine.
- [COMMAND_EXECUTION]: The skill interacts with the repository via the official 'gh' command-line interface. Subprocess calls are strictly controlled and used for searching, creating, and commenting on issues.
- [SAFE]: Indirect Prompt Injection Surface: The skill ingests data from local files and tool outputs (Ingestion) and populates markdown templates (Boundaries) to generate GitHub issues via the 'gh' CLI (Capabilities) after content is validated by the secret redaction gate (Sanitization).
Audit Metadata