next
Next
The user knows where they are. They just need directions.
On Activation
- Quick scan - unfinished work, logical next steps, pending items, quick wins
- Check TaskList - if tasks exist, surface unblocked ones
- Present 2-4 options via
AskUserQuestion- they tap, you execute
No status tables. No reconstruction. Just options.
Option Quality
Start each option with a verb. Be concrete.
Good:
- "Finish the auth refactor in
api.ts" - "Run tests and fix failures"
- "Add error handling to the new endpoint"
Bad:
- "Continue working" (vague)
- "Maybe look at tests?" (uncertain)
- "We could do several things..." (narrative)
Example Output
AskUserQuestion:
question: "What's next?"
header: "Next"
options:
- label: "Add tests for UserService"
description: "Cover the new authentication methods"
- label: "Wire up the frontend"
description: "Connect login form to the new endpoint"
- label: "Clean up TODO comments"
description: "Quick win - 3 items flagged earlier"
When Context is Empty
If there's nothing to go on:
- Ask what they'd like to work on
- Offer to explore the codebase
- Suggest reviewing recent git activity
Anti-patterns
- Status tables (that's /debrief)
- Reconstructing what happened (that's /debrief)
- More than 4 options (decision paralysis)
- Fewer than 2 options (not useful)
- Making them type when they could tap
Enter next mode now. Scan context, check TaskList, present 2-4 options via AskUserQuestion. No preamble.
More from corygabrielsen/skills
mission-control
Coordinate complex multi-step work using task graphs and parallel background agents. Use when work requires decomposition, delegation, and long-running operations that may survive context compaction.
11loop-codex-review
Automated code review loop with progressive reasoning levels. Runs n parallel Codex reviews (configurable via -n), Claude addresses issues, climbs from low→xhigh reasoning until fixed point (all n clean). Human approval at each iteration.
11review-pr
Thorough, educational PR review process. Gathers context, explains changes, identifies issues systematically, and drafts constructive feedback collaboratively before submitting.
11orthogonalize-pr
Identify and separate orthogonal (independent) change sets within a decomposed PR. Creates branches for each set that could theoretically be reviewed or merged separately.
10decontextualize
Course-correct when artifacts overfit to conversation context. Invoke when examples, names, or language are too specific to current work.
10loop-review-skill-until-fixed-point
Iterate /review-skill on a target until fixed point. Runs review passes until all reviewers return NO ISSUES.
9