---
name: "gsd-spec-phase"
description: "Clarify WHAT a phase delivers with ambiguity scoring; produces a SPEC.md before discuss-phase."
metadata:
  short-description: "Clarify WHAT a phase delivers with ambiguity scoring; produces a SPEC.md before discuss-phase."
---

<codex_skill_adapter>
## A. Skill Invocation
- This skill is invoked by mentioning `$gsd-spec-phase`.
- Treat all user text after `$gsd-spec-phase` as `{{GSD_ARGS}}`.
- If no arguments are present, treat `{{GSD_ARGS}}` as empty.

## B. AskUserQuestion → request_user_input Mapping
GSD workflows use `AskUserQuestion` (Claude Code syntax). Translate to Codex `request_user_input`:

Parameter mapping:
- `header` → `header`
- `question` → `question`
- Options formatted as `"Label" — description` → `{label: "Label", description: "description"}`
- Generate `id` from header: lowercase, replace spaces with underscores

Batched calls:
- `AskUserQuestion([q1, q2])` → single `request_user_input` with multiple entries in `questions[]`

Multi-select workaround:
- Codex has no `multiSelect`. Use sequential single-selects, or present a numbered freeform list asking the user to enter comma-separated numbers.

Execute mode fallback:
- When `request_user_input` is rejected or unavailable, you MUST stop and present the questions as a plain-text numbered list, then wait for the user's reply. Do NOT pick a default and continue (#3018).
- You may only proceed without a user answer when one of these is true:
  (a) the invocation included an explicit non-interactive flag (`--auto` or `--all`),
  (b) the user has explicitly approved a specific default for this question, or
  (c) the workflow's documented contract says defaults are safe (e.g. autonomous lifecycle paths).
- Do NOT write workflow artifacts (CONTEXT.md, DISCUSSION-LOG.md, PLAN.md, checkpoint files) until the user has answered the plain-text questions or one of (a)-(c) above applies. Surfacing the questions and waiting is the correct response — silently defaulting and writing artifacts is the #3018 failure mode.

## C. Task() → spawn_agent Mapping
GSD workflows use `Task(...)` (Claude Code syntax). Translate to Codex collaboration tools:

Direct mapping:
- `Task(subagent_type="X", prompt="Y")` → `spawn_agent(agent_type="X", message="Y")`
- `Task(model="...")` → omit. `spawn_agent` has no inline `model` parameter;
  GSD embeds the resolved per-agent model directly into each agent's `.toml`
  at install time so `model_overrides` from `.planning/config.json` and
  `~/.gsd/defaults.json` are honored automatically by Codex's agent router.
- Resolved `reasoning_effort="low|medium|high|xhigh"` (`xhigh` is a GSD/Codex tier, not a generic runtime enum) → pass `reasoning_effort`
  to `spawn_agent` when the runtime/tool supports it. Omit missing, empty,
  inherited, or unsupported values; do not invent one-off effort literals in
  workflow prose.
- `fork_context: false` by default — GSD agents load their own context via `<files_to_read>` blocks
- `Task(isolation="worktree")` / `Agent(isolation="worktree")` → no direct Codex mapping.
  Codex `spawn_agent` does not create or bind a git worktree automatically.
  Workflows that require this isolation must fail closed or use an explicit
  manual worktree protocol before spawning (#3360).

Spawn restriction:
- Codex restricts `spawn_agent` to cases where the user has explicitly
  requested sub-agents. When automatic spawning is not permitted, do the
  work inline in the current agent rather than attempting to force a spawn.

Parallel fan-out:
- Spawn multiple agents → collect agent IDs → `wait(ids)` for all to complete

Result parsing:
- Look for structured markers in agent output: `CHECKPOINT`, `PLAN COMPLETE`, `SUMMARY`, etc.
- `close_agent(id)` after collecting results from each agent
</codex_skill_adapter>

<objective>
Clarify phase requirements through structured Socratic questioning with quantitative ambiguity scoring.

**Position in workflow:** `spec-phase → discuss-phase → plan-phase → execute-phase → verify`

**How it works:**
1. Load phase context (PROJECT.md, REQUIREMENTS.md, ROADMAP.md, STATE.md)
2. Scout the codebase — understand current state before asking questions
3. Run Socratic interview loop (up to 6 rounds, rotating perspectives)
4. Score ambiguity across 4 weighted dimensions after each round
5. Gate: ambiguity ≤ 0.20 AND all dimensions meet minimums → write SPEC.md
6. Commit SPEC.md — discuss-phase picks it up automatically on next run

**Output:** `{phase_dir}/{padded_phase}-SPEC.md` — falsifiable requirements that lock "what/why" before discuss-phase handles "how"
</objective>

<execution_context>
@$HOME/.codex/get-shit-done/workflows/spec-phase.md
@$HOME/.codex/get-shit-done/templates/spec.md
</execution_context>

<runtime_note>
**Copilot (VS Code):** Use `vscode_askquestions` wherever this workflow calls `AskUserQuestion`. They are equivalent.
</runtime_note>

<context>
Phase number: {{GSD_ARGS}} (required)

**Flags:**
- `--auto` — Skip interactive questions; the agent selects recommended defaults and writes SPEC.md
- `--text` — Use plain-text numbered lists instead of TUI menus (required for `/rc` remote sessions)

Context files are resolved in-workflow using `init phase-op`.
</context>

<process>
Execute end-to-end.

**MANDATORY:** Read the workflow file BEFORE taking any action. The workflow contains the complete step-by-step process including the Socratic interview loop, ambiguity scoring gate, and SPEC.md generation. Do not improvise from the objective summary above.
</process>

<success_criteria>
- Codebase scouted for current state before questioning begins
- All 4 ambiguity dimensions scored after each interview round
- Gate passed: ambiguity ≤ 0.20 AND all dimension minimums met
- SPEC.md written with falsifiable requirements, explicit boundaries, and acceptance criteria
- SPEC.md committed atomically
- User knows they can now run $gsd-discuss-phase which will load SPEC.md automatically
</success_criteria>