The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs actions using a local binary called lin. It uses Bash(lin *) to execute subcommands for managing Linear's workspace data.
[EXTERNAL_DOWNLOADS]: Facilitates downloading file attachments directly from Linear's official CDN ("uploads.linear.app"). This is a standard feature for retrieving shared workspace assets.
[PROMPT_INJECTION]: The skill processes content from Linear (issues, comments, documents), which constitutes an indirect prompt injection surface for untrusted data provided by other workspace users.
Ingestion points: Data enters the context through list and get commands in SKILL.md.
Boundary markers: No delimiters are present to isolate retrieved content from agent instructions.
Capability inventory: Includes capabilities for workspace modification and file system downloads.
Sanitization: Content from the Linear API is not validated or sanitized before processing.
[SAFE]: The skill follows security best practices by instructing the agent not to directly access or extract Linear API keys, instead delegating authentication to the underlying CLI tool.

lin