spk-run-next

What it does

Drives the canonical spec-kitty next --mission <handle> control loop for mission advancement and routes the returned decision — query, step, blocked, decision_required, or terminal — to the correct next action.

When to reach for it

Use it whenever you want to advance an active mission, ask what to do next, or recover from a runtime decision. It is the everyday entry point for driving a mission forward one step at a time from inside your agent harness.

Invocation

• Query state: spec-kitty next --mission <handle> --json
• Advance state: spec-kitty next --agent <name> --mission <handle> --result <success|failed|blocked>

The skill's own frontmatter declares no fixed trigger phrase; its detailed legacy alias documents the natural-language requests that route here — "run the next step," "what should runtime do next," "advance the mission," "what is the next task," "continue the workflow," "what step comes next."

Flow

1. query — inspect only; do not execute a prompt or mark a result.
2. step — execute the generated prompt file.
3. decision_required — answer with --answer, --result, --agent, and --decision-id when multiple decisions are pending.
4. blocked — fix guard failures before retrying.
5. terminal — route to spk-gate-accept.

Legacy alias

For detailed runtime semantics — the decision algorithm, WP iteration logic, guard primitives, the prompt-file contract, and known issues — use spec-kitty-runtime-next when it is available in your agent's skill set.