ticket
Pass
Audited by Gen Agent Trust Hub on May 13, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill consists exclusively of instructional text and does not include any executable scripts or binary files, significantly reducing the potential for traditional code-based exploits.
- [SAFE]: The skill employs a human-in-the-loop pattern by requiring explicit user approval before creating tickets via the Linear MCP tool, which effectively mitigates risks associated with processing untrusted user input.
- [SAFE]: All external interactions are confined to the Linear MCP tool for workspace metadata retrieval and ticket creation, with no evidence of arbitrary network access or data exfiltration.
- [PROMPT_INJECTION]: The skill's ingestion of free-form user descriptions constitutes an architectural surface for indirect prompt injection; however, the requirement for manual review maintains a safe posture. Ingestion point: User description (Step 1). Boundary markers: Absent. Capability inventory: Linear MCP (listing teams, projects, labels, and creating issues). Sanitization: Absent.
Audit Metadata