gsd-antigravity-kit 2.0.0 → 2.1.0

package/.agent/skills/gsd/references/workflows/discuss-phase.md

@@ -113,6 +113,15 @@ Phase: "API documentation"
 
 <answer_validation>
 **IMPORTANT: Answer validation** — After every AskUserQuestion call, check if the response is empty or whitespace-only. If so:
+
+**Exception — "Other" with empty text:** If the user selected "Other" (or "Chat more") and the response body is empty or whitespace-only, this is NOT an empty answer — it is a signal that the user wants to type freeform input. In this case:
+1. Output a single plain-text line: "What would you like to discuss?"
+2. STOP generating. Do not call any tools. Do not output any further text.
+3. Wait for the user's next message.
+4. After receiving their message, reflect it back and continue.
+Do NOT retry the AskUserQuestion or generate more questions when "Other" is selected with empty text.
+
+**All other empty responses:** If the response is empty or whitespace-only (and the user did NOT select "Other"):
 1. Retry the question once with the same parameters
 2. If still empty, present the options as a plain-text numbered list and ask the user to type their choice number
 Never proceed with an empty answer.
@@ -125,7 +134,7 @@ cannot forward TUI menu selections back to the host.
 
 Enable text mode:
 - Per-session: pass `--text` flag to any command (e.g., `/gsd-discuss-phase --text`)
-- Per-project: `gsd-tools config-set workflow.text_mode true`
+- Per-project: `gsd-sdk query config-set workflow.text_mode true`
 
 Text mode applies to ALL workflows in the session, not just discuss-phase.
 </answer_validation>
@@ -138,9 +147,9 @@ Text mode applies to ALL workflows in the session, not just discuss-phase.
 Phase number from argument (required).
 
 ```bash
-.agent/skills/gsd/bin/gsd-tools.cjs" init phase-op "${PHASE}")
+INIT=$(gsd-sdk query init.phase-op "${PHASE}")
 if [[ "$INIT" == @file:* ]]; then INIT=$(cat "${INIT#@file:}"); fi
-.agent/skills/gsd/bin/gsd-tools.cjs" agent-skills gsd-advisor 2>/dev/null)
+AGENT_SKILLS_ADVISOR=$(gsd-sdk query agent-skills gsd-advisor 2>/dev/null)
 ```
 
 Parse JSON for: `commit_docs`, `phase_found`, `phase_dir`, `phase_number`, `phase_name`, `phase_slug`, `padded_phase`, `has_research`, `has_context`, `has_plans`, `has_verification`, `plan_count`, `roadmap_exists`, `planning_exists`, `response_language`.
@@ -162,6 +171,13 @@ Exit workflow.
 - Read and execute @references/workflows/discuss-phase-power.md end-to-end
 - Do not continue with the steps below
 
+**All mode** — If `--all` is present in ARGUMENTS:
+- In `present_gray_areas`: auto-select ALL gray areas without asking the user (skips the AskUserQuestion selection step)
+- Discussion for each area proceeds fully interactively (user drives the conversation for every area)
+- Does NOT auto-advance to plan-phase afterward — use `--chain` or `--auto` if you want auto-advance
+- Log: `[--all] Auto-selected all gray areas: [list area names].`
+- This is the "discuss everything" shortcut: skip the selection friction, keep full interactive control
+
 **Auto mode** — If `--auto` is present in ARGUMENTS:
 - In `check_existing`: auto-select "Skip" (if context exists) or continue without prompting (if no context/plans)
 - In `present_gray_areas`: auto-select ALL gray areas without asking the user
@@ -196,7 +212,30 @@ This step cannot be skipped. Before proceeding to `check_existing` or any other
 
 Write these answers inline before continuing. If a blocking anti-pattern cannot be answered from the context in `.continue-here.md`, stop and ask the user for clarification.
 
-**If no `.continue-here.md` exists, or no `blocking` rows are found:** Proceed directly to `check_existing`.
+**If no `.continue-here.md` exists, or no `blocking` rows are found:** Proceed directly to `check_spec`.
+</step>
+
+<step name="check_spec">
+Check if a SPEC.md (from `/gsd-spec-phase`) exists for this phase. SPEC.md locks requirements before implementation decisions — if present, this discussion focuses on HOW to implement, not WHAT to build.
+
+```bash
+ls ${phase_dir}/*-SPEC.md 2>/dev/null | grep -v AI-SPEC | head -1 || true
+```
+
+**If SPEC.md is found:**
+1. Read the SPEC.md file.
+2. Count the number of requirements (numbered items in the `## Requirements` section).
+3. Display:
+   ```
+   Found SPEC.md — {N} requirements locked. Focusing on implementation decisions.
+   ```
+4. Set internal flag `spec_loaded = true`.
+5. Store the requirements, boundaries, and acceptance criteria from SPEC.md as `<locked_requirements>` — these flow directly into CONTEXT.md without re-asking.
+6. Continue to `check_existing`.
+
+**If no SPEC.md is found:** Continue to `check_existing` with `spec_loaded = false` (default behavior unchanged).
+
+**Note:** SPEC.md files named `AI-SPEC.md` (from `/gsd-ai-integration-phase`) are excluded — those serve a different purpose.
 </step>
 
 <step name="check_existing">
@@ -309,9 +348,40 @@ Structure the extracted information:
 </prior_decisions>
 ```
 
+**Step 4: Load spike/sketch findings (if they exist)**
+```bash
+# Check for spike/sketch findings skills (project-local)
+SPIKE_FINDINGS=$(ls ./.antigravity/skills/spike-findings-*/SKILL.md 2>/dev/null | head -1)
+SKETCH_FINDINGS=$(ls ./.antigravity/skills/sketch-findings-*/SKILL.md 2>/dev/null | head -1)
+
+# Also check for raw spikes/sketches not yet wrapped up
+RAW_SPIKES=$(ls .planning/spikes/MANIFEST.md 2>/dev/null)
+RAW_SKETCHES=$(ls .planning/sketches/MANIFEST.md 2>/dev/null)
+```
+
+If spike/sketch findings skills exist, read their SKILL.md and reference files. Extract:
+- **Validated patterns** — what was proven to work (use these, don't re-explore)
+- **Landmines** — what was proven NOT to work (avoid these)
+- **Constraints** — hard limits discovered (rate limits, API gaps, library limitations)
+- **Design decisions** — winning visual directions, CSS patterns, layout choices
+
+Add to `<prior_decisions>`:
+```
+## From Spike Experiments
+- [Validated pattern or constraint from spike findings]
+
+## From Design Sketches
+- [Design decision or visual direction from sketch findings]
+```
+
+If raw spikes/sketches exist but no findings skill, note in output:
+```
+⚠ Unpackaged spikes/sketches detected — run `/gsd-spike-wrap-up` or `/gsd-sketch-wrap-up` to make findings available to planning agents.
+```
+
 **Usage in subsequent steps:**
-- `analyze_phase`: Skip gray areas already decided in prior phases
-- `present_gray_areas`: Annotate options with prior decisions ("You chose X in Phase 5")
+- `analyze_phase`: Skip gray areas already decided in prior phases or validated by spikes/sketches
+- `present_gray_areas`: Annotate options with prior decisions ("You chose X in Phase 5") and spike/sketch findings ("Spike 002 validated this approach")
 - `discuss_areas`: Pre-fill answers or flag conflicts ("This contradicts Phase 3 — same here or different?")
 
 **If no prior context exists:** Continue without — this is expected for early phases.
@@ -322,7 +392,7 @@ Check if any pending todos are relevant to this phase's scope. Surfaces backlog
 
 **Load and match todos:**
 ```bash
-.agent/skills/gsd/bin/gsd-tools.cjs" todo match-phase "${PHASE_NUMBER}")
+TODO_MATCHES=$(gsd-sdk query todo.match-phase "${PHASE_NUMBER}")
 ```
 
 Parse JSON for: `todo_count`, `matches[]` (each with `file`, `title`, `area`, `score`, `reasons`).
@@ -421,6 +491,12 @@ Analyze the phase to identify gray areas worth discussing. **Use both `prior_dec
    - These are **pre-answered** — don't re-ask unless this phase has conflicting needs
    - Note applicable prior decisions for use in presentation
 
+2b. **SPEC.md awareness** — If `spec_loaded = true` (SPEC.md was found in `check_spec`):
+   - The `<locked_requirements>` from SPEC.md are pre-answered: Goal, Boundaries, Constraints, Acceptance Criteria.
+   - Do NOT generate gray areas about WHAT to build or WHY — those are locked.
+   - Only generate gray areas about HOW to implement: technical approach, library choices, UX/UI patterns, interaction details, error handling style.
+   - When presenting gray areas, include a note: "Requirements are locked by SPEC.md — discussing implementation decisions only."
+
 3. **Gray areas by category** — For each relevant category (UI, UX, Behavior, Empty States, Content), identify 1-2 specific ambiguities that would change implementation. **Annotate with code context where relevant** (e.g., "You already have a Card component" or "No existing pattern for this").
 
 4. **Skip assessment** — If no meaningful gray areas exist (pure infrastructure, clear-cut implementation, or all already decided in prior phases), the phase may not need discussion.
@@ -447,11 +523,39 @@ Check if advisor mode should activate:
 
 3. Resolve model for advisor agents:
    ```bash
-.agent/skills/gsd/bin/gsd-tools.cjs" resolve-model gsd-advisor-researcher --raw)
+   ADVISOR_MODEL=$(gsd-sdk query resolve-model gsd-advisor-researcher --raw)
    ```
 
 If ADVISOR_MODE is false, skip all advisor-specific steps — workflow proceeds with existing conversational flow unchanged.
 
+**User Profile Language Detection:**
+
+Check USER-PROFILE.md for communication preferences that indicate a non-technical product owner:
+
+```bash
+PROFILE_CONTENT=$(cat "C:/projects/GSD-Antigravity/.antigravity/get-shit-done/USER-PROFILE.md" 2>/dev/null || true)
+```
+
+Set NON_TECHNICAL_OWNER = true if ANY of the following are present in USER-PROFILE.md:
+- `learning_style: guided`
+- The word `jargon` appears in a `frustration_triggers` section
+- `explanation_depth: practical-detailed` (without a technical modifier)
+- `explanation_depth: high-level`
+
+NON_TECHNICAL_OWNER = false if USER-PROFILE.md does not exist or none of the above signals are present.
+
+When NON_TECHNICAL_OWNER is true, reframe gray area labels and descriptions in product-outcome language before presenting them to the user. Preserve the same underlying decision — only change the framing:
+- Technical implementation term → outcome the user will experience
+  - "Token architecture" → "Color system: which approach prevents the dark theme from flashing white on open"
+  - "CSS variable strategy" → "Theme colors: how your brand colors stay consistent in both light and dark mode"
+  - "Component API surface area" → "How the building blocks connect: how tightly coupled should these parts be"
+  - "Caching strategy: SWR vs React Query" → "Loading speed: should screens show saved data right away or wait for fresh data"
+- All decisions stay the same. Only the question language adapts.
+
+This reframing applies to:
+1. Gray area labels and descriptions in `present_gray_areas`
+2. Advisor research rationale rewrites in `advisor_research` synthesis
+
 **Output your analysis internally, then present to user.**
 
 Example analysis for "Post Feed" phase (with code and prior context):
@@ -485,7 +589,7 @@ We'll clarify HOW to implement this.
 - [Decision from Phase M that applies here]
 ```
 
-**If `--auto`:** Auto-select ALL gray areas. Log: `[auto] Selected all gray areas: [list area names].` Skip the AskUserQuestion below and continue directly to discuss_areas with all areas selected.
+**If `--auto` or `--all`:** Auto-select ALL gray areas. Log: `[--auto/--all] Selected all gray areas: [list area names].` Skip the AskUserQuestion below and continue directly to discuss_areas with all areas selected.
 
 **Otherwise, use AskUserQuestion (multiSelect: true):**
 - header: "Discuss"
@@ -581,6 +685,7 @@ After user selects gray areas in present_gray_areas, spawn parallel research age
       If agent returned too many, trim least viable. If too few, accept as-is.
    d. Rewrite rationale paragraph to weave in project context and ongoing discussion context that the agent did not have access to
    e. If agent returned only 1 option, convert from table format to direct recommendation: "Standard approach for {area}: {option}. {rationale}"
+   f. **If NON_TECHNICAL_OWNER is true:** After completing steps a–e, apply a plain language rewrite to the rationale paragraph. Replace implementation-level terms with outcome descriptions the user can reason about without technical context. The table option names may also be rewritten in plain language if they are implementation terms — the Recommendation column value and the table structure remain intact. Do not remove detail; translate it. Example: "SWR uses stale-while-revalidate to serve cached responses immediately" → "This approach shows you something right away, then quietly updates in the background — users see data instantly."
 
 4. Store synthesized tables for use in discuss_areas.
 
@@ -721,7 +826,7 @@ In `--auto` mode, the discuss step MUST complete in a **single pass**. After wri
 
 Check the pass cap from config:
 ```bash
-.agent/skills/gsd/bin/gsd-tools.cjs" config-get workflow.max_discuss_passes 2>/dev/null || echo "3")
+MAX_PASSES=$(gsd-sdk query config-get workflow.max_discuss_passes 2>/dev/null || echo "3")
 ```
 
 If you have already written and committed CONTEXT.md, the discuss step is complete. Move on.
@@ -870,6 +975,12 @@ mkdir -p ".planning/phases/${padded_phase}-${phase_slug}"
 
 **File location:** `${phase_dir}/${padded_phase}-CONTEXT.md`
 
+**SPEC.md integration** — If `spec_loaded = true`:
+- Add a `<spec_lock>` section immediately after `<domain>` (see template below).
+- Add the SPEC.md file to `<canonical_refs>` with note "Locked requirements — MUST read before planning".
+- Do NOT duplicate requirements text from SPEC.md into `<decisions>` — agents read SPEC.md directly.
+- The `<decisions>` section contains only implementation decisions from this discussion.
+
 **Structure the content by what was discussed:**
 
 ```markdown
@@ -885,6 +996,19 @@ mkdir -p ".planning/phases/${padded_phase}-${phase_slug}"
 
 </domain>
 
+[If spec_loaded = true, insert this section:]
+<spec_lock>
+## Requirements (locked via SPEC.md)
+
+**{N} requirements are locked.** See `{padded_phase}-SPEC.md` for full requirements, boundaries, and acceptance criteria.
+
+Downstream agents MUST read `{padded_phase}-SPEC.md` before planning or implementing. Requirements are not duplicated here.
+
+**In scope (from SPEC.md):** [copy the "In scope" bullet list from SPEC.md Boundaries]
+**Out of scope (from SPEC.md):** [copy the "Out of scope" bullet list from SPEC.md Boundaries]
+
+</spec_lock>
+
 <decisions>
 ## Implementation Decisions
 
@@ -993,7 +1117,7 @@ Created: .planning/phases/${PADDED_PHASE}-${SLUG}/${PADDED_PHASE}-CONTEXT.md
 
 ---
 
-## ▶ Next Up
+## ▶ Next Up — [${PROJECT_CODE}] ${PROJECT_TITLE}
 
 **Phase ${PHASE}: [Name]** — [Goal from ROADMAP.md]
 
@@ -1067,7 +1191,7 @@ rm -f "${phase_dir}/${padded_phase}-DISCUSS-CHECKPOINT.json"
 Commit phase context and discussion log:
 
 ```bash
-.agent/skills/gsd/bin/gsd-tools.cjs" commit "docs(${padded_phase}): capture phase context" --files "${phase_dir}/${padded_phase}-CONTEXT.md" "${phase_dir}/${padded_phase}-DISCUSSION-LOG.md"
+gsd-sdk query commit "docs(${padded_phase}): capture phase context" "${phase_dir}/${padded_phase}-CONTEXT.md" "${phase_dir}/${padded_phase}-DISCUSSION-LOG.md"
 ```
 
 Confirm: "Committed: docs(${padded_phase}): capture phase context"
@@ -1077,7 +1201,7 @@ Confirm: "Committed: docs(${padded_phase}): capture phase context"
 Update STATE.md with session info:
 
 ```bash
-.agent/skills/gsd/bin/gsd-tools.cjs" state record-session \
+gsd-sdk query state.record-session \
   --stopped-at "Phase ${PHASE} context gathered" \
   --resume-file "${phase_dir}/${padded_phase}-CONTEXT.md"
 ```
@@ -1085,29 +1209,29 @@ Update STATE.md with session info:
 Commit STATE.md:
 
 ```bash
-.agent/skills/gsd/bin/gsd-tools.cjs" commit "docs(state): record phase ${PHASE} context session" --files .planning/STATE.md
+gsd-sdk query commit "docs(state): record phase ${PHASE} context session" .planning/STATE.md
 ```
 </step>
 
 <step name="auto_advance">
 Check for auto-advance trigger:
 
-1. Parse `--auto` and `--chain` flags from $ARGUMENTS
+1. Parse `--auto` and `--chain` flags from $ARGUMENTS. Note: --all is NOT an auto-advance trigger — it only affects area selection. A session with `--all` but without `--auto` or `--chain` returns to manual next-steps after discussion completes.
 2. **Sync chain flag with intent** — if user invoked manually (no `--auto` and no `--chain`), clear the ephemeral chain flag from any previous interrupted `--auto` chain. This does NOT touch `workflow.auto_advance` (the user's persistent settings preference):
    ```bash
    if [[ ! "$ARGUMENTS" =~ --auto ]] && [[ ! "$ARGUMENTS" =~ --chain ]]; then
-.agent/skills/gsd/bin/gsd-tools.cjs" config-set workflow._auto_chain_active false 2>/dev/null
+     gsd-sdk query config-set workflow._auto_chain_active false 2>/dev/null
    fi
    ```
 3. Read both the chain flag and user preference:
    ```bash
-.agent/skills/gsd/bin/gsd-tools.cjs" config-get workflow._auto_chain_active 2>/dev/null || echo "false")
-.agent/skills/gsd/bin/gsd-tools.cjs" config-get workflow.auto_advance 2>/dev/null || echo "false")
+   AUTO_CHAIN=$(gsd-sdk query config-get workflow._auto_chain_active 2>/dev/null || echo "false")
+   AUTO_CFG=$(gsd-sdk query config-get workflow.auto_advance 2>/dev/null || echo "false")
    ```
 
 **If `--auto` or `--chain` flag present AND `AUTO_CHAIN` is not true:** Persist chain flag to config (handles direct usage without new-project):
 ```bash
-.agent/skills/gsd/bin/gsd-tools.cjs" config-set workflow._auto_chain_active true
+gsd-sdk query config-set workflow._auto_chain_active true
 ```
 
 **If `--auto` flag present OR `--chain` flag present OR `AUTO_CHAIN` is true OR `AUTO_CFG` is true:**

package/.agent/skills/gsd/references/workflows/do.md

@@ -11,6 +11,8 @@ Read all files referenced by the invoking prompt's execution_context before star
 <step name="validate">
 **Check for input.**
 
+
+**Text mode (`workflow.text_mode: true` in config or `--text` flag):** Set `TEXT_MODE=true` if `--text` is present in `$ARGUMENTS` OR `text_mode` from init JSON is `true`. When TEXT_MODE is active, replace every `AskUserQuestion` call with a plain-text numbered list and ask the user to type their choice number. This is required for non-Antigravity runtimes (OpenAI Codex, Gemini CLI, etc.) where `AskUserQuestion` is not available.
 If `$ARGUMENTS` is empty, ask via AskUserQuestion:
 
 ```
@@ -24,7 +26,7 @@ Wait for response before continuing.
 **Check if project exists.**
 
 ```bash
-.agent/skills/gsd/bin/gsd-tools.cjs" state load 2>/dev/null)
+INIT=$(gsd-sdk query state.load 2>/dev/null)
 ```
 
 Track whether `.planning/` exists — some routes require it, others don't.
@@ -40,6 +42,10 @@ Evaluate `$ARGUMENTS` against these routing rules. Apply the **first matching**
 | Starting a new project, "set up", "initialize" | `/gsd-new-project` | Needs full project initialization |
 | Mapping or analyzing an existing codebase | `/gsd-map-codebase` | Codebase discovery |
 | A bug, error, crash, failure, or something broken | `/gsd-debug` | Needs systematic investigation |
+| Spiking, "test if", "will this work", "experiment", "prove this out", validate feasibility | `/gsd-spike` | Throwaway experiment to validate feasibility |
+| Sketching, "mockup", "what would this look like", "prototype the UI", "design this", explore visual direction | `/gsd-sketch` | Throwaway HTML mockups to explore design |
+| Wrapping up spikes, "package the spikes", "consolidate spike findings" | `/gsd-spike-wrap-up` | Package spike findings into reusable skill |
+| Wrapping up sketches, "package the designs", "consolidate sketch findings" | `/gsd-sketch-wrap-up` | Package sketch findings into reusable skill |
 | Exploring, researching, comparing, or "how does X work" | `/gsd-research-phase` | Domain research before planning |
 | Discussing vision, "how should X look", brainstorming | `/gsd-discuss-phase` | Needs context gathering |
 | A complex task: refactoring, migration, multi-file architecture, system redesign | `/gsd-add-phase` | Needs a full phase with plan/build cycle |
@@ -54,7 +60,7 @@ Evaluate `$ARGUMENTS` against these routing rules. Apply the **first matching**
 | Completing a milestone, shipping, releasing | `/gsd-complete-milestone` | Milestone lifecycle |
 | A specific, actionable, small task (add feature, fix typo, update config) | `/gsd-quick` | Self-contained, single executor |
 
-**Requires `.planning/` directory:** All routes except `/gsd-new-project`, `/gsd-map-codebase`, `/gsd-help`, and `/gsd-join-discord`. If the project doesn't exist and the route requires it, suggest `/gsd-new-project` first.
+**Requires `.planning/` directory:** All routes except `/gsd-new-project`, `/gsd-map-codebase`, `/gsd-spike`, `/gsd-sketch`, `/gsd-help`, and `/gsd-join-discord`. If the project doesn't exist and the route requires it, suggest `/gsd-new-project` first.
 
 **Ambiguity handling:** If the text could reasonably match multiple routes, ask the user via AskUserQuestion with the top 2-3 options. For example:
 

package/.agent/skills/gsd/references/workflows/docs-update.md

@@ -14,9 +14,9 @@ Valid GSD subagent types (use exact names — do not fall back to 'general-purpo
 Load docs-update context:
 
 ```bash
-.agent/skills/gsd/bin/gsd-tools.cjs" docs-init)
+INIT=$(gsd-sdk query docs-init)
 if [[ "$INIT" == @file:* ]]; then INIT=$(cat "${INIT#@file:}"); fi
-.agent/skills/gsd/bin/gsd-tools.cjs" agent-skills gsd-doc-writer 2>/dev/null)
+AGENT_SKILLS=$(gsd-sdk query agent-skills gsd-doc-writer 2>/dev/null)
 ```
 
 Extract from init JSON:
@@ -84,6 +84,8 @@ Assemble the complete doc queue from always-on docs plus conditional docs from c
 If CONTRIBUTING.md is in the conditional queue AND does NOT appear in the `existing_docs` array from init JSON:
 
 1. If `--force` is present in `$ARGUMENTS`: skip this check, include CONTRIBUTING.md in the queue.
+
+**Text mode (`workflow.text_mode: true` in config or `--text` flag):** Set `TEXT_MODE=true` if `--text` is present in `$ARGUMENTS` OR `text_mode` from init JSON is `true`. When TEXT_MODE is active, replace every `AskUserQuestion` call with a plain-text numbered list and ask the user to type their choice number. This is required for non-Antigravity runtimes (OpenAI Codex, Gemini CLI, etc.) where `AskUserQuestion` is not available.
 2. Otherwise, use AskUserQuestion to confirm:
 
 ```
@@ -1052,7 +1054,7 @@ Only run this step if `commit_docs` is `true` from the init JSON. If `commit_doc
 Assemble the list of files that were actually generated (do not include files that failed or were skipped):
 
 ```bash
-.agent/skills/gsd/bin/gsd-tools.cjs" commit "docs: generate project documentation" \
+gsd-sdk query commit "docs: generate project documentation" \
   --files README.md docs/ARCHITECTURE.md docs/CONFIGURATION.md docs/GETTING-STARTED.md docs/DEVELOPMENT.md docs/TESTING.md
 # Append any conditional docs that were generated:
 # --files ... docs/API.md docs/DEPLOYMENT.md CONTRIBUTING.md

package/.agent/skills/gsd/references/workflows/eval-review.md

@@ -0,0 +1,155 @@
+<purpose>
+Retroactive audit of an implemented AI phase's evaluation coverage. Standalone command that works on any GSD-managed AI phase. Produces a scored EVAL-REVIEW.md with gap analysis and remediation plan.
+
+Use after /gsd-execute-phase to verify that the evaluation strategy from AI-SPEC.md was actually implemented. Mirrors the pattern of /gsd-ui-review and /gsd-validate-phase.
+</purpose>
+
+<required_reading>
+@references/docs/ai-evals.md
+</required_reading>
+
+<process>
+
+## 0. Initialize
+
+```bash
+INIT=$(gsd-sdk query init.phase-op "${PHASE_ARG}")
+if [[ "$INIT" == @file:* ]]; then INIT=$(cat "${INIT#@file:}"); fi
+```
+
+Parse: `phase_dir`, `phase_number`, `phase_name`, `phase_slug`, `padded_phase`, `commit_docs`.
+
+```bash
+AUDITOR_MODEL=$(gsd-sdk query resolve-model gsd-eval-auditor 2>/dev/null | jq -r '.model' 2>/dev/null || true)
+```
+
+Display banner:
+```
+━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
+ GSD ► EVAL AUDIT — PHASE {N}: {name}
+━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
+```
+
+## 1. Detect Input State
+
+```bash
+SUMMARY_FILES=$(ls "${PHASE_DIR}"/*-SUMMARY.md 2>/dev/null)
+AI_SPEC_FILE=$(ls "${PHASE_DIR}"/*-AI-SPEC.md 2>/dev/null | head -1)
+EVAL_REVIEW_FILE=$(ls "${PHASE_DIR}"/*-EVAL-REVIEW.md 2>/dev/null | head -1)
+```
+
+**State A** — AI-SPEC.md + SUMMARY.md exist: Full audit against spec
+**State B** — SUMMARY.md exists, no AI-SPEC.md: Audit against general best practices
+**State C** — No SUMMARY.md: Exit — "Phase {N} not executed. Run /gsd-execute-phase {N} first."
+
+
+**Text mode (`workflow.text_mode: true` in config or `--text` flag):** Set `TEXT_MODE=true` if `--text` is present in `$ARGUMENTS` OR `text_mode` from init JSON is `true`. When TEXT_MODE is active, replace every `AskUserQuestion` call with a plain-text numbered list and ask the user to type their choice number. This is required for non-Antigravity runtimes (OpenAI Codex, Gemini CLI, etc.) where `AskUserQuestion` is not available.
+**If `EVAL_REVIEW_FILE` non-empty:** Use AskUserQuestion:
+- header: "Existing Eval Review"
+- question: "EVAL-REVIEW.md already exists for Phase {N}."
+- options:
+  - "Re-audit — run fresh audit"
+  - "View — display current review and exit"
+
+If "View": display file, exit.
+If "Re-audit": continue.
+
+**If State B (no AI-SPEC.md):** Warn:
+```
+No AI-SPEC.md found for Phase {N}.
+Audit will evaluate against general AI eval best practices rather than a phase-specific plan.
+Consider running /gsd-ai-integration-phase {N} before implementation next time.
+```
+Continue (non-blocking).
+
+## 2. Gather Context Paths
+
+Build file list for auditor:
+- AI-SPEC.md (if exists — the planned eval strategy)
+- All SUMMARY.md files in phase dir
+- All PLAN.md files in phase dir
+
+## 3. Spawn gsd-eval-auditor
+
+```
+◆ Spawning eval auditor...
+```
+
+Build prompt:
+
+```markdown
+references/agents/gsd-eval-auditor.md for instructions.
+
+<objective>
+Conduct evaluation coverage audit of Phase {phase_number}: {phase_name}
+{If AI-SPEC exists: "Audit against AI-SPEC.md evaluation plan."}
+{If no AI-SPEC: "Audit against general AI eval best practices."}
+</objective>
+
+<files_to_read>
+- {summary_paths}
+- {plan_paths}
+- {ai_spec_path if exists}
+</files_to_read>
+
+<input>
+ai_spec_path: {ai_spec_path or "none"}
+phase_dir: {phase_dir}
+phase_number: {phase_number}
+phase_name: {phase_name}
+padded_phase: {padded_phase}
+state: {A or B}
+</input>
+```
+
+Spawn as Task with model `AUDITOR_MODEL`.
+
+## 4. Parse Auditor Result
+
+Read the written EVAL-REVIEW.md. Extract:
+- `overall_score`
+- `verdict` (PRODUCTION READY | NEEDS WORK | SIGNIFICANT GAPS | NOT IMPLEMENTED)
+- `critical_gap_count`
+
+## 5. Display Summary
+
+```
+━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
+ GSD ► EVAL AUDIT COMPLETE — PHASE {N}: {name}
+━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
+
+◆ Score: {overall_score}/100
+◆ Verdict: {verdict}
+◆ Critical Gaps: {critical_gap_count}
+◆ Output: {eval_review_path}
+
+{If PRODUCTION READY:}
+  Next step: /gsd-plan-phase (next phase) or deploy
+
+{If NEEDS WORK:}
+  Address critical gaps in EVAL-REVIEW.md, then re-run /gsd-eval-review {N}
+
+{If SIGNIFICANT GAPS or NOT IMPLEMENTED:}
+  Review AI-SPEC.md evaluation plan. Critical eval dimensions are not implemented.
+  Do not deploy until gaps are addressed.
+```
+
+## 6. Commit
+
+**If `commit_docs` is true:**
+```bash
+git add "${EVAL_REVIEW_FILE}"
+git commit -m "docs({phase_slug}): add EVAL-REVIEW.md — score {overall_score}/100 ({verdict})"
+```
+
+</process>
+
+<success_criteria>
+- [ ] Phase execution state detected correctly
+- [ ] AI-SPEC.md presence handled (with or without)
+- [ ] gsd-eval-auditor spawned with correct context
+- [ ] EVAL-REVIEW.md written (by auditor)
+- [ ] Score and verdict displayed to user
+- [ ] Appropriate next steps surfaced based on verdict
+- [ ] Committed if commit_docs enabled
+</success_criteria>