The Agent Skills Directory

[COMMAND_EXECUTION]: The skill operates by executing the linear CLI binary. It uses specific commands like linear status, linear authorize-url, linear list-tools, and linear call-tool to interact with Linear's services. This is the intended core functionality of the skill and follows standard CLI-based agent integration patterns.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests untrusted data from an external platform (Linear). Maliciously crafted content within Linear issues or comments could be interpreted as instructions by the agent when it reads them.
Ingestion points: Data is brought into the agent context through methods such as issues.read, comments.read, projects.read, and documents.read as defined in manifest.yaml.
Boundary markers: The instructions in SKILL.md do not specify the use of delimiters or clear boundary markers to isolate data retrieved from Linear from the agent's core instructions.
Capability inventory: The skill possesses command execution capabilities through the linear call-tool interface, which can be used to create or update issues, projects, and comments.
Sanitization: There is no evidence of sanitization, filtering, or validation of the content retrieved from Linear before it is presented to the agent.

linear