The Agent Skills Directory

[PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it ingests untrusted data from the Linear API (issue titles, descriptions, and comments) and has the capability to perform mutations.
Ingestion points: Commands such as linear-cli i get, linear-cli i list, and linear-cli cm list pull data from external Linear issues into the agent's context.
Boundary markers: The instructions require the agent to state whether each command is read-only or mutating and mandates the use of --dry-run for bulk operations.
Capability inventory: The skill allows the agent to create/update issues, projects, and webhooks using linear-cli i create, linear-cli i update, and linear-cli wh create.
Sanitization: The skill includes explicit warnings against GraphQL injection and instructs the agent to use GraphQL variables (-v) rather than string interpolation.
[COMMAND_EXECUTION]: The skill relies on executing shell commands to interact with the Linear CLI, jq, and the GitHub gh CLI. It uses subshell expansion and piping to manage data flows.
[EXTERNAL_DOWNLOADS]: The skill mentions external installation and update paths including linear-cli update and cargo install linear-cli. These are standard package management and update operations for the tool described in the skill.

use-linear-cli