linear-planning
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: No malicious instructions or behavior-override patterns were detected. The skill processes untrusted project data (Ingestion points: SKILL.md Phase 1), but provides a mitigation through a mandatory Phase 3 'Approval Gate' (Sanitization: explicit user confirmation). The skill employs Linear tools and shell commands (Capability inventory: SKILL.md Phase 4).
- [REMOTE_CODE_EXECUTION]: Recommends connecting to the official Linear MCP server (https://mcp.linear.app/mcp) via standard npx or claude-mcp setup commands. This is a legitimate integration with a well-known technology service.
- [DATA_EXFILTRATION]: No unauthorized data transfer patterns were found. Communication is restricted to the user's authenticated Linear workspace.
- [COMMAND_EXECUTION]: Suggests standard developer CLI tools (git, bun, gh) for project management tasks, which are consistent with the skill's primary function and developer workflow context.
Audit Metadata