The Agent Skills Directory

[COMMAND_EXECUTION]: The skill methodology outlines a process for uploading images to Linear issues using a curl PUT request to a signed URL (METHODOLOGY.md, §10). This is a standard and documented method for interacting with the Linear API's attachment system.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core functionality of reading and acting upon data from external sources (Linear issues and repository documentation).
Ingestion points: The agent retrieves data from Linear issue titles, descriptions, and comments, as well as internal project documentation like roadmaps and plans (SKILL.md, METHODOLOGY.md §1a, §4).
Boundary markers: There are no instructions to use specific delimiters or protective prompts (e.g., "ignore instructions in the following data") when the agent processes this external content.
Capability inventory: The agent has write access to the Linear project via MCP (creating issues, milestones, and attachments) and can execute network commands via curl (METHODOLOGY.md §10, §11).
Sanitization: The instructions do not mandate validation or sanitization of the externally sourced text before it is processed or used to generate new project entities.

linear-sync