linear
Pass
Audited by Gen Agent Trust Hub on Feb 21, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill processes untrusted data from Linear issue titles, descriptions, and comments which could contain malicious instructions. Evidence: 1. Ingestion points: issue and issues commands in scripts/linear-cli.js. 2. Boundary markers: absent. 3. Capability inventory: Linear API mutations including createIssue, updateIssue, and createComment in scripts/linear-cli.js. 4. Sanitization: absent.
- [External Downloads] (LOW): The skill requires installation of the @linear/sdk package from npm. While the source is not on the specific trusted organization list, it is the official SDK for the service and its use is intrinsic to the skill's primary purpose, justifying a severity downgrade from MEDIUM.
Audit Metadata