linear-teamlead
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill mandates that no credentials be included in Linear profiles, preventing accidental credential exposure during the orchestration process.
- [SAFE]: A role-based verification mechanism is established where agents must match the authenticated Linear actor to their assigned role before publication, mitigating the risk of unauthorized actions.
- [SAFE]: The skill requires internal metadata such as unit keys and worker IDs to be kept in isolated scratch directories outside the repository, preventing sensitive environment data from leaking into public trackers or reports.
- [INDIRECT_PROMPT_INJECTION]: The skill handles data from Linear issues, representing an indirect prompt injection surface. 1. Ingestion points: Linear issue titles and IDs are ingested via the integration context block. 2. Boundary markers: Structured fields such as LINEAR ISSUE and LINEAR ROLE are used as delimiters. 3. Capability inventory: The skill allows publishing artifacts, updating status, and changing metadata. 4. Sanitization: Not explicitly performed; risk is mitigated by the identity verification and role constraints. This is evaluated as safe given the implementation of strict integration boundaries and the requirement for agents to verify actors before performing any write operations.
Audit Metadata