github-issue-create
Pass
Audited by Gen Agent Trust Hub on May 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates within expected parameters for a GitHub management utility, using the official
ghCLI. It includes strict procedural safeguards, such as requiring human approval for all write operations, which ensures that no actions are taken autonomously. - [INDIRECT_PROMPT_INJECTION]: The skill's potential for indirect prompt injection via external GitHub comments was evaluated and found to be mitigated by its design.
- Ingestion points: Context is gathered from repository issues and comments using
ghcommands (SKILL.md). - Boundary markers: Explicit separation between the context-gathering 'Draft' phase and the 'Publish' phase (SKILL.md).
- Capability inventory: Includes creating issues, labels, and modifying native relationships via the GitHub API (references/gh-commands.md).
- Sanitization: Mandatory manual review and explicit user approval before execution of the final plan (SKILL.md).
Audit Metadata