The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it retrieves data from Linear issues (titles, descriptions, comments) which may contain malicious instructions that influence agent behavior. (1) Ingestion points: Data enters the context via tools like LINEAR_SEARCH_ISSUES, LINEAR_GET_LINEAR_ISSUE, and LINEAR_LIST_LINEAR_ISSUES. (2) Boundary markers: The instructions lack delimiters or explicit directives to ignore instructions within the ingested data. (3) Capability inventory: The skill can perform various write operations and execute arbitrary GraphQL queries via LINEAR_RUN_QUERY_OR_MUTATION. (4) Sanitization: No evidence of input validation or escaping of external content is provided.
[EXTERNAL_DOWNLOADS]: The skill requires the user to configure a third-party MCP server at https://rube.app/mcp. This directs workspace interaction and authentication through external infrastructure (Composio/Rube) not included in the trusted vendors list.

linear-automation