linear-finish-install
Pass
Audited by Gen Agent Trust Hub on Jul 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill performs expected configuration tasks by writing behavioral guidelines to standard agent configuration files such as ~/.claude/CLAUDE.md and ~/.codex/AGENTS.md.
- [SAFE]: It utilizes official Model Context Protocol (MCP) tools for Linear (e.g., list_teams, get_user) and provides links to well-known, trusted domains including linear.app, cursor.com, and claude.ai.
- [SAFE]: AUTH.md specifically instructs the agent not to invent unofficial API wrappers or raw curl commands, which is a security-positive constraint that prevents the use of unauthenticated or malicious endpoints.
- [SAFE]: No evidence of prompt injection, data exfiltration, obfuscation, or unauthorized command execution was detected in any of the provided files.
Audit Metadata