The Agent Skills Directory

[Indirect Prompt Injection] (LOW): The skill is susceptible to indirect prompt injection due to its interaction with external data sources.
Ingestion points: The skill reads external data via linear issue list, linear issue view, and git log.
Boundary markers: There are no delimiters or instructions provided to the agent to ignore potentially malicious content embedded in issue descriptions or commit messages.
Capability inventory: The skill allows the agent to execute state changes (linear issue update) and create pull requests (linear issue pr).
Sanitization: No evidence of sanitization or validation of the retrieved external content before it is processed by the agent.
[Data Exposure & Exfiltration] (SAFE): No hardcoded credentials or access to sensitive local files (like SSH keys) were detected. The environment variable LINEAR_ISSUE_SORT is used for non-sensitive configuration.
[Unverifiable Dependencies] (SAFE): References the linear CLI tool via Homebrew, which is a standard and trusted package manager for macOS developer environments.
[Command Execution] (SAFE): The commands used (linear, git) are standard development tools and are used according to their intended primary purpose within the context of issue tracking.

linear