configure-issue-tracker
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources and local files, creating a potential surface for indirect prompt injection, although the risk is mitigated by user review.
- Ingestion points: Content is read from
docs/agents/issue-tracker.md,AGENTS.md, and the output of thegit remote -vcommand. - Boundary markers: The skill does not implement specific delimiters (e.g., XML tags or block quotes) to isolate ingested data from instructions.
- Capability inventory: The skill's primary capability is writing and patching local Markdown files. It does not possess network exfiltration capabilities or the ability to execute arbitrary code.
- Sanitization: The skill incorporates a mandatory confirmation step, requiring the agent to show proposed differences to the user and obtain approval before applying any changes, which acts as a manual sanitization layer.
Audit Metadata