what-to-work

name	what-to-work
description	Use when you do not know what to work on next, the user asks 'what should I work on' / '뭐하지', or a session starts over an existing STV backlog. Routes to what-we-have-to-work when ANY unfinished scenario exists, or plan-new-task when the backlog is empty.

What To Work

Goal

Provide clear, user-confirmable next work options. Scan docs/*/trace.md for unfinished scenarios. If ANY unfinished scenario exists, route to stv:what-we-have-to-work. Only when the backlog is empty (or the user explicitly asks for new-feature planning) route to stv:plan-new-task.

Decision Gate (MANDATORY)

Read ${CLAUDE_PLUGIN_ROOT}/prompts/decision-gate.md and apply it to every decision in this workflow.

Sizing Rubric: read ${CLAUDE_PLUGIN_ROOT}/prompts/decision-gate.md (single source — do not duplicate the table here). Sizes below (tiny/small/medium/large/xlarge) refer to expected code change (added + deleted).

Workflow

1. Scan trace files

   • Glob for docs/*/trace.md in the project
   • Read each trace.md's Implementation Status table
   • Collect scenarios where Status != "Complete"
   • Estimate size for each unfinished scenario

2. Decide if work exists

   • Execution-eligible: at least one unfinished scenario exists — ANY size, including tiny/medium tails. Leftover work is never abandoned behind new-feature planning.
   • Empty backlog: zero unfinished scenarios, or the user explicitly bypassed the backlog.

3. Route

   • If execution-eligible → Skill(skill="stv:what-we-have-to-work") to propose 1-3 bundles
   • If empty backlog (or explicit bypass) → Skill(skill="stv:plan-new-task") to propose new features

4. Present next action

   • State which route you are using and why
   • Ask for any missing context only if it blocks routing

Error Handling

• Missing or malformed Implementation Status table → report "trace format mismatch: {file}" and stop; never silently fall through to planning.
• Inconsistent status labels across traces → instruct the maintainer to normalize to the Scenario | Trace | Tests | Verify | Status schema before routing.
• Empty glob result → verify docs/ was actually scanned before concluding the backlog is empty; a glob/tooling failure is not an empty backlog.

Output Template

Route: what-we-have-to-work

Trace scan complete: {N} features found, {M} unfinished scenarios
Unfinished: {M} scenarios ({sizes})
Total estimated work: {tier}
Route: what-we-have-to-work
Next step: I'll bundle scenarios into up to three options for you to pick.

Route: plan-new-task

Trace scan complete: {summary}
Backlog empty: 0 unfinished scenarios.
Route: plan-new-task
Next step: I'll propose new features based on completed work and project context.

Integration

• Use stv:what-we-have-to-work whenever at least one unfinished scenario exists (any size)
• Use stv:plan-new-task only when all scenarios are complete or the user explicitly requests new-feature planning
• After user selection, follow stv:do-work for execution