codely-plan-create-github
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill performs legitimate project management tasks using established tools (GitHub CLI). It includes a critical human-in-the-loop step, requiring user approval before any modifications or creations are made to the repository's issues. Data handling is limited to the context of GitHub issues within the current working directory repository.
- [INDIRECT_PROMPT_INJECTION]: While the skill ingests external content from GitHub issues, which is a potential surface for indirect prompt injection, it is well-mitigated by the workflow design.
- Ingestion points: Reading content from the parent GitHub issue via
gh issue view <parent>inSKILL.md. - Boundary markers: The instructions explicitly demand user approval ("Propose first, get approval, then create the issues") and follow strict guidelines in
resources/plan-guidelines.md. - Capability inventory: The skill uses
gh issue create,gh issue edit, andgh apito manage issues. It does not perform arbitrary code execution or external network calls outside of GitHub. - Sanitization: The skill relies on model interpretation and user verification rather than programmatic sanitization of the issue body.
Audit Metadata