The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or unauthorized data exfiltration attempts were detected. The skill demonstrates professional engineering practices, including the use of schema-driven communication and explicit validation steps.
[COMMAND_EXECUTION]: The skill executes a provided local TypeScript script (scripts/validate_marked_comments.ts) to validate its output against structured YAML schemas. This is a secure and standard practice for ensuring that machine-readable outputs are valid and have not been malformed or manipulated.
[PROMPT_INJECTION]: The skill processes untrusted input from Linear issues and comments, which represents a surface for indirect prompt injection. However, this is effectively mitigated by the skill's reliance on rigid YAML schemas for all critical instructions and the mandatory use of a validation tool to verify inputs before they can influence the agent's planning logic.

linear-refine