work-issue
Pass
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill adopts a high-security posture by explicitly instructing the agent to treat all content retrieved from GitHub (titles, bodies, comments) as untrusted input, effectively mitigating indirect prompt injection risks.
- [SAFE]: It mandates strict validation for repository identifiers and issue numbers, ensuring they contain only safe, expected characters before use in system commands.
- [SAFE]: The workflow requires safe shell practices, such as quoting variables and avoiding the direct interpolation of raw external data into shell commands or script logic.
- [SAFE]: Branch naming includes a secure slugification process with a character whitelist (lowercase alphanumeric and hyphens) to prevent injection attacks using malicious issue titles.
- [SAFE]: All network and system interactions are scoped to legitimate GitHub operations using standard, well-known tooling (
gh,git), with no signs of malicious exfiltration or unauthorized access.
Audit Metadata