to-tickets
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external URLs, issue bodies, and comments to generate ticket titles and descriptions. This creates a surface where malicious instructions embedded in a specification or issue could attempt to influence the agent's behavior.
- Ingestion points: The skill reads the body and comments of issues from URLs or identifiers provided as arguments, as well as local specification files.
- Boundary markers: The skill does not explicitly use delimiters or instructions to ignore embedded prompts when reading external content.
- Capability inventory: The skill has the ability to read files, write local files to the
.scratch/directory, and interact with external issue trackers (GitHub, Linear) via the configured environment. - Sanitization: No explicit sanitization or filtering of the ingested content is described in the instructions, although the "Quiz the user" step provides a manual verification gate before persistence.
Audit Metadata