to-tickets
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to fetch and read content from external references such as specification paths, issue numbers, or URLs. This data is then used to generate task tickets. If these external sources contain adversarial instructions (e.g., hidden prompts to modify the codebase maliciously), the agent might follow them while processing the context.
- Ingestion points: The 'Gather context' section in
SKILL.mdinstructs the agent to fetch and read the full body and comments of external references. - Boundary markers: There are no specific instructions or delimiters mentioned to isolate the external content or warn the agent against executing embedded commands.
- Capability inventory: The skill possesses capabilities to explore the codebase (read files), write local files (
tickets.md), and interact with external issue trackers (GitHub, Linear). - Sanitization: No sanitization or validation logic is defined for the content retrieved from external sources.
Audit Metadata