to-issues
Pass
Audited by Gen Agent Trust Hub on Jul 20, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted external content such as plans, specifications, PRDs, and issue comments to generate tracker issues.
- Ingestion points: Reads issue bodies, comments, and project files as context for drafting new issues (
SKILL.md). - Boundary markers: The provided issue templates do not utilize specific delimiters or instructions to isolate or ignore potentially malicious embedded content within the processed data.
- Capability inventory: The skill allows the agent to publish new issues directly to a project issue tracker (
SKILL.md). - Sanitization: There is no evidence of explicit sanitization or validation of the input text; however, the risk is mitigated by a mandatory human review step before publication and the platform-level restriction on model invocation.
Audit Metadata