菜单
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.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use when the user's request is ambiguous — vague asks, polysemous instructions, unclear scope — BEFORE committing to a feature. Two-track loop (one question at a time + parallel codebase exploration) producing a Context Brief that becomes the SSOT for the follow-up work.
Use when code behaves differently from expectations — 'why does this happen', 'find the bug', 'follow the callstack', symptom reports even without the word debugging. Blackbox methodology: AS-IS/TO-BE confirmation, callstack tracing recorded in a CWD-relative trace file, red-green fix.
Use when a trace.md with unfinished scenarios exists and the user wants autonomous implementation — 'continue the work', 'implement the traced feature', or after a bundle was selected. Accepts an explicit bundle contract (trace_path + scenario_ids); scans for ready scenarios only when no contract is given.
Use when the problem is vague, the domain unfamiliar, multiple approaches are possible, or the user says 'investigate first' / 'before we spec this'. Read-only stance (not a workflow): no code changes, no mandatory artifacts; exits to spec/debug/clarify/direct implementation when insight crystallizes.
Use when the user gives a vague or high-level feature request that needs decomposition into implementable work — new ideas, '~만들어줘' with implicit requirements, features with multiple possible approaches. NOT for specific 1-2 file changes, obvious bug fixes, quick questions, or when spec/trace already exist (use stv:do-work).
Use when the trace backlog has ZERO unfinished scenarios (or the user explicitly asks for new feature ideas) and the project needs its next feature proposed. Hands the chosen idea to new-task for spec + trace.
| 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. |
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.
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).
Scan trace files
docs/*/trace.md in the projectDecide if work exists
Route
Skill(skill="stv:what-we-have-to-work") to propose 1-3 bundlesSkill(skill="stv:plan-new-task") to propose new featuresPresent next action
Scenario | Trace | Tests | Verify | Status schema before routing.docs/ was actually scanned before concluding the backlog is empty; a glob/tooling failure is not an empty backlog.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.
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.
stv:what-we-have-to-work whenever at least one unfinished scenario exists (any size)stv:plan-new-task only when all scenarios are complete or the user explicitly requests new-feature planningstv:do-work for execution