linear
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves issue descriptions, comments, and document contents from the Linear API, which may contain adversarial instructions from external users.
- Ingestion points: GraphQL queries in
SKILL.md(e.g.,issueSearch,issue(id:),documents) fetch content from Linear. - Boundary markers: The instructions do not define specific delimiters or warnings to prevent the agent from obeying instructions embedded in the retrieved data.
- Capability inventory: The skill allows mutations (creation and updates) to Linear issues and projects via
curlcommands in the terminal. - Sanitization: There are no instructions for the agent to sanitize or validate the external content before processing it.
- [SAFE]: Interaction with Linear is limited to the official API endpoint (
https://api.linear.app/graphql) and the official MCP server (https://mcp.linear.app/mcp). - [SAFE]: Authentication is handled correctly via a
LINEAR_API_KEYenvironment variable, avoiding the use of hardcoded credentials.
Audit Metadata