linear-planning
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external codebase files (such as READMEs, package.json, and specifications) and user-provided descriptions to generate Linear tickets. This creates a surface for indirect prompt injection where instructions embedded in the analyzed code could attempt to influence the agent's ticket generation.
- Ingestion points: Reads repository source files, architecture docs, and requirement specifications in Phase 1 (SKILL.md).
- Capability inventory: Linear MCP tools for creating and updating issues;
git,bun, andghCLI tools for workflow implementation (SKILL.md, Phase 4). - Boundary markers: No specific delimiters or warnings for the file-analysis phase are defined.
- Sanitization: The skill explicitly requires a mandatory Approval Gate (Phase 3) where the user must explicitly approve the full issue tree before any creation tools are called. This human review process effectively mitigates the risk of automated obedience to injected instructions.
Audit Metadata