The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it retrieves and processes untrusted data from the Linear platform that could contain malicious instructions.
Ingestion points: Untrusted data enters the agent context through commands like linear issue view, linear issue list, linear issue comment list, and linear project view which fetch titles, descriptions, and comments from the Linear API.
Boundary markers: The skill does not provide instructions or delimiters to isolate fetched content or warn the agent to ignore embedded instructions.
Capability inventory: The agent has the capability to execute shell commands (linear), modify the local environment (export LINEAR_API_KEY), and perform destructive actions like linear issue delete.
Sanitization: There is no evidence of sanitization, filtering, or validation of the content retrieved from Linear before it is processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill instructs the agent to download and install the Linear CLI from external sources managed by the author.
It installs the @injectivelabs/linear-cli package via NPM or fetches binaries directly from github.com/InjectiveLabs/linear-cli/releases.
[COMMAND_EXECUTION]: The skill relies on the execution of shell commands to interact with the Linear service and manage local configuration.
Commands include linear auth login, linear config for generating configuration files, and various issue management commands like linear issue start and linear issue update.

linear-cli