linear
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines patterns for querying issue descriptions, comments, and attachments, which are sources of untrusted external data.
- Ingestion points: SKILL.md contains GraphQL query templates for retrieving issue details and comments.
- Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions for the retrieved content.
- Capability inventory: The skill utilizes the linear_graphql tool for write operations and instructs the use of curl via the shell.
- Sanitization: There are no instructions for sanitizing or validating the data ingested from Linear.
- [COMMAND_EXECUTION]: The skill explicitly instructs the agent to use the curl command-line tool for file uploads.
- Evidence: The 'File upload' section in SKILL.md directs the agent to 'PUT file bytes to uploadUrl with the returned headers (use curl)'.
Audit Metadata