The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the bd CLI tool for managing issues within the project's tracker. These commands are consistent with the skill's stated engineering workflow.
[PROMPT_INJECTION]: The skill processes untrusted input (plans/specs) to generate tasks. Ingestion points: user-provided documentation in SKILL.md. Boundary markers: absent. Capability inventory: issue creation and retrieval via the bd tool. Sanitization: none specified. This represents a surface for indirect prompt injection where malicious input could influence task generation, though the impact is restricted to the issue tracker.

to-features