The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes standard Git commands such as git status --short and git log -1 --oneline to verify the state of the repository before and after subagent tasks. These are routine operations for a development orchestration tool.
[INDIRECT_PROMPT_INJECTION]: The skill reads content from local ticket files provided by the user (specifically lines like Difficulty: and Blocked by:) to determine execution order and feasibility. While this represents an ingestion point for untrusted data, the skill is explicitly configured with disable-model-invocation: true, requiring manual trigger by the user, which mitigates the risk of autonomous exploitation.
[SAFE]: The skill implements safety 'gates' such as checking for a clean working tree before starting and stopping execution if a task is marked as 'Heavy' difficulty, ensuring human oversight for complex changes.

delegate-tickets