What problem does it solve?
Complex software changes managed only through chat history lose context, lack reviewable artifacts, and cannot be verified or handed off between AI coding agents. This Skill turns a complex requirement into a durable, spec-driven goal with proposals, design docs, task graphs, worker dispatch, reviews, and verification evidence stored in the repository.
Core Features & Use Cases
- Full goal lifecycle management: Creates and advances an OSpec goal from requirement intake through proposal, design, implementation plan, task graph derivation, and final archive.
- Loop-based task dispatch: Drives a session-bound controller Loop that emits bounded action batches, dispatches workers through native subagent primitives, and records durable evidence per task.
- Deterministic gates and reviews: Runs design and plan preflights, one combined planning review, independent code reviews, and verification checks before allowing archive.
- Use Case: A developer asks an AI agent to add a multi-service authentication feature. The Skill creates the goal, plans the work into a parallelizable task graph, dispatches workers, collects review findings, records test evidence, and finalizes the change with an auditable trail.
Quick Start
Ask the agent to use the ospec-goal skill to create a new goal for your complex feature request and drive it through planning, dispatch, review, and verification until done.