Extract implementation decisions that downstream agents need. Analyze the phase to identify gray areas, let the user choose what to discuss, then deep-dive each selected area until satisfied. You are a thinking partner, not an interviewer. The user is the visionary — you are the builder. Your job is to capture decisions that will guide research and planning, not to figure out implementation yourself. **CONTEXT.md feeds into:** 1. **gsd-phase-researcher** — Reads CONTEXT.md to know WHAT to research - "User wants card-based layout" → researcher investigates card component patterns - "Infinite scroll decided" → researcher looks into virtualization libraries 2. **gsd-planner** — Reads CONTEXT.md to know WHAT decisions are locked - "Pull-to-refresh on mobile" → planner includes that in task specs - "Claude's Discretion: loading skeleton" → planner can decide approach **Your job:** Capture decisions clearly enough that downstream agents can act on them without asking the user again. **Not your job:** Figure out HOW to implement. That's what research and planning do with the decisions you capture. **User = founder/visionary. Claude = builder.** The user knows: - How they imagine it working - What it should look/feel like - What's essential vs nice-to-have - Specific behaviors or references they have in mind The user doesn't know (and shouldn't be asked): - Codebase patterns (researcher reads the code) - Technical risks (researcher identifies these) - Implementation approach (planner figures this out) - Success metrics (inferred from the work) Ask about vision and implementation choices. Capture decisions for downstream agents. **CRITICAL: No scope creep.** The phase boundary comes from ROADMAP.md and is FIXED. Discussion clarifies HOW to implement what's scoped, never WHETHER to add new capabilities. **Allowed (clarifying ambiguity):** - "How should posts be displayed?" (layout, density, info shown) - "What happens on empty state?" (within the feature) - "Pull to refresh or manual?" (behavior choice) **Not allowed (scope creep):** - "Should we also add comments?" (new capability) - "What about search/filtering?" (new capability) - "Maybe include bookmarking?" (new capability) **The heuristic:** Does this clarify how we implement what's already in the phase, or does it add a new capability that could be its own phase? **When user suggests scope creep:** ``` "[Feature X] would be a new capability — that's its own phase. Want me to note it for the roadmap backlog? For now, let's focus on [phase domain]." ``` Capture the idea in a "Deferred Ideas" section. Don't lose it, don't act on it. Gray areas are **implementation decisions the user cares about** — things that could go multiple ways and would change the result. **How to identify gray areas:** 1. **Read the phase goal** from ROADMAP.md 2. **Understand the domain** — What kind of thing is being built? - Something users SEE → visual presentation, interactions, states matter - Something users CALL → interface contracts, responses, errors matter - Something users RUN → invocation, output, behavior modes matter - Something users READ → structure, tone, depth, flow matter - Something being ORGANIZED → criteria, grouping, handling exceptions matter 3. **Generate phase-specific gray areas** — Not generic categories, but concrete decisions for THIS phase **Don't use generic category labels** (UI, UX, Behavior). Generate specific gray areas: ``` Phase: "User authentication" → Session handling, Error responses, Multi-device policy, Recovery flow Phase: "Organize photo library" → Grouping criteria, Duplicate handling, Naming convention, Folder structure Phase: "CLI for database backups" → Output format, Flag design, Progress reporting, Error recovery Phase: "API documentation" → Structure/navigation, Code examples depth, Versioning approach, Interactive elements ``` **The key question:** What decisions would change the outcome that the user should weigh in on? **Claude handles these (don't ask):** - Technical implementation details - Architecture patterns - Performance optimization - Scope (roadmap defines this) **Express path available:** If you already have a PRD or acceptance criteria document, use `/gsd:plan-phase {phase} --prd path/to/prd.md` to skip this discussion and go straight to planning. Phase number from argument (required). ```bash INIT=$(node "$HOME/.claude/get-shit-done/bin/gsd-tools.cjs" init phase-op "${PHASE}") ``` 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`. **If `phase_found` is false:** ``` Phase [X] not found in roadmap. Use /gsd:progress to see available phases. ``` Exit workflow. **If `phase_found` is true:** Continue to check_existing. Check if CONTEXT.md already exists using `has_context` from init. ```bash ls ${phase_dir}/*-CONTEXT.md 2>/dev/null ``` **If exists:** Use AskUserQuestion: - header: "Context" - question: "Phase [X] already has context. What do you want to do?" - options: - "Update it" — Review and revise existing context - "View it" — Show me what's there - "Skip" — Use existing context as-is If "Update": Load existing, continue to analyze_phase If "View": Display CONTEXT.md, then offer update/skip If "Skip": Exit workflow **If doesn't exist:** Check `has_plans` and `plan_count` from init. **If `has_plans` is true:** Use AskUserQuestion: - header: "Plans exist" - question: "Phase [X] already has {plan_count} plan(s) created without user context. Your decisions here won't affect existing plans unless you replan." - options: - "Continue and replan after" — Capture context, then run /gsd:plan-phase {X} to replan - "View existing plans" — Show plans before deciding - "Cancel" — Skip discuss-phase If "Continue and replan after": Continue to analyze_phase. If "View existing plans": Display plan files, then offer "Continue" / "Cancel". If "Cancel": Exit workflow. **If `has_plans` is false:** Continue to scout_codebase. Lightweight scan of existing code to inform gray area identification and discussion. Uses ~10% context — acceptable for an interactive session. **Step 1: Check for existing codebase maps** ```bash ls .planning/codebase/*.md 2>/dev/null ``` **If codebase maps exist:** Read the most relevant ones (CONVENTIONS.md, STRUCTURE.md, STACK.md based on phase type). Extract: - Reusable components/hooks/utilities - Established patterns (state management, styling, data fetching) - Integration points (where new code would connect) Skip to Step 3 below. **Step 2: If no codebase maps, do targeted grep** Extract key terms from the phase goal (e.g., "feed" → "post", "card", "list"; "auth" → "login", "session", "token"). ```bash # Find files related to phase goal terms grep -rl "{term1}\|{term2}" src/ app/ --include="*.ts" --include="*.tsx" --include="*.js" --include="*.jsx" 2>/dev/null | head -10 # Find existing components/hooks ls src/components/ 2>/dev/null ls src/hooks/ 2>/dev/null ls src/lib/ src/utils/ 2>/dev/null ``` Read the 3-5 most relevant files to understand existing patterns. **Step 3: Build internal codebase_context** From the scan, identify: - **Reusable assets** — existing components, hooks, utilities that could be used in this phase - **Established patterns** — how the codebase does state management, styling, data fetching - **Integration points** — where new code would connect (routes, nav, providers) - **Creative options** — approaches the existing architecture enables or constrains Store as internal `` for use in analyze_phase and present_gray_areas. This is NOT written to a file — it's used within this session only. Analyze the phase to identify gray areas worth discussing. **Use codebase_context from scout step to ground the analysis.** **Read the phase description from ROADMAP.md and determine:** 1. **Domain boundary** — What capability is this phase delivering? State it clearly. 2. **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"). 3. **Skip assessment** — If no meaningful gray areas exist (pure infrastructure, clear-cut implementation), the phase may not need discussion. **Output your analysis internally, then present to user.** Example analysis for "Post Feed" phase (with code context): ``` Domain: Displaying posts from followed users Existing: Card component (src/components/ui/Card.tsx), useInfiniteQuery hook, Tailwind CSS Gray areas: - UI: Layout style (cards vs timeline vs grid) — Card component exists with shadow/rounded variants - UI: Information density (full posts vs previews) — no existing density patterns - Behavior: Loading pattern (infinite scroll vs pagination) — useInfiniteQuery already set up - Empty State: What shows when no posts exist — EmptyState component exists in ui/ - Content: What metadata displays (time, author, reactions count) ``` Present the domain boundary and gray areas to user. **First, state the boundary:** ``` Phase [X]: [Name] Domain: [What this phase delivers — from your analysis] We'll clarify HOW to implement this. (New capabilities belong in other phases.) ``` **Then use AskUserQuestion (multiSelect: true):** - header: "Discuss" - question: "Which areas do you want to discuss for [phase name]?" - options: Generate 3-4 phase-specific gray areas, each with: - "[Specific area]" (label) — concrete, not generic - [1-2 questions this covers + code context annotation] (description) - **Highlight the recommended choice with brief explanation why** **Code context annotations:** When the scout found relevant existing code, annotate the gray area description: ``` ☐ Layout style — Cards vs list vs timeline? (You already have a Card component with shadow/rounded variants. Reusing it keeps the app consistent.) ``` **Do NOT include a "skip" or "you decide" option.** User ran this command to discuss — give them real choices. **Examples by domain (with code context):** For "Post Feed" (visual feature): ``` ☐ Layout style — Cards vs list vs timeline? (Card component exists with variants) ☐ Loading behavior — Infinite scroll or pagination? (useInfiniteQuery hook available) ☐ Content ordering — Chronological, algorithmic, or user choice? ☐ Post metadata — What info per post? Timestamps, reactions, author? ``` For "Database backup CLI" (command-line tool): ``` ☐ Output format — JSON, table, or plain text? Verbosity levels? ☐ Flag design — Short flags, long flags, or both? Required vs optional? ☐ Progress reporting — Silent, progress bar, or verbose logging? ☐ Error recovery — Fail fast, retry, or prompt for action? ``` For "Organize photo library" (organization task): ``` ☐ Grouping criteria — By date, location, faces, or events? ☐ Duplicate handling — Keep best, keep all, or prompt each time? ☐ Naming convention — Original names, dates, or descriptive? ☐ Folder structure — Flat, nested by year, or by category? ``` Continue to discuss_areas with selected areas. For each selected area, conduct a focused discussion loop. **Philosophy: 4 questions, then check.** Ask 4 questions per area before offering to continue or move on. Each answer often reveals the next question. **For each area:** 1. **Announce the area:** ``` Let's talk about [Area]. ``` 2. **Ask 4 questions using AskUserQuestion:** - header: "[Area]" (max 12 chars — abbreviate if needed) - question: Specific decision for this area - options: 2-3 concrete choices (AskUserQuestion adds "Other" automatically), with the recommended choice highlighted and brief explanation why - **Annotate options with code context** when relevant: ``` "How should posts be displayed?" - Cards (reuses existing Card component — consistent with Messages) - List (simpler, would be a new pattern) - Timeline (needs new Timeline component — none exists yet) ``` - Include "You decide" as an option when reasonable — captures Claude discretion - **Context7 for library choices:** When a gray area involves library selection (e.g., "magic links" → query next-auth docs) or API approach decisions, use `mcp__context7__*` tools to fetch current documentation and inform the options. Don't use Context7 for every question — only when library-specific knowledge improves the options. 3. **After 4 questions, check:** - header: "[Area]" (max 12 chars) - question: "More questions about [area], or move to next?" - options: "More questions" / "Next area" If "More questions" → ask 4 more, then check again If "Next area" → proceed to next selected area If "Other" (free text) → interpret intent: continuation phrases ("chat more", "keep going", "yes", "more") map to "More questions"; advancement phrases ("done", "move on", "next", "skip") map to "Next area". If ambiguous, ask: "Continue with more questions about [area], or move to the next area?" 4. **After all initially-selected areas complete:** - Summarize what was captured from the discussion so far - AskUserQuestion: - header: "Done" - question: "We've discussed [list areas]. Which gray areas remain unclear?" - options: "Explore more gray areas" / "I'm ready for context" - If "Explore more gray areas": - Identify 2-4 additional gray areas based on what was learned - Return to present_gray_areas logic with these new areas - Loop: discuss new areas, then prompt again - If "I'm ready for context": Proceed to write_context **Question design:** - Options should be concrete, not abstract ("Cards" not "Option A") - Each answer should inform the next question - If user picks "Other", receive their input, reflect it back, confirm **Scope creep handling:** If user mentions something outside the phase domain: ``` "[Feature] sounds like a new capability — that belongs in its own phase. I'll note it as a deferred idea. Back to [current area]: [return to current question]" ``` Track deferred ideas internally. Create CONTEXT.md capturing decisions made. **Find or create phase directory:** Use values from init: `phase_dir`, `phase_slug`, `padded_phase`. If `phase_dir` is null (phase exists in roadmap but no directory): ```bash mkdir -p ".planning/phases/${padded_phase}-${phase_slug}" ``` **File location:** `${phase_dir}/${padded_phase}-CONTEXT.md` **Structure the content by what was discussed:** ```markdown # Phase [X]: [Name] - Context **Gathered:** [date] **Status:** Ready for planning ## Phase Boundary [Clear statement of what this phase delivers — the scope anchor] ## Implementation Decisions ### [Category 1 that was discussed] - [Decision or preference captured] - [Another decision if applicable] ### [Category 2 that was discussed] - [Decision or preference captured] ### Claude's Discretion [Areas where user said "you decide" — note that Claude has flexibility here] ## Existing Code Insights ### Reusable Assets - [Component/hook/utility]: [How it could be used in this phase] ### Established Patterns - [Pattern]: [How it constrains/enables this phase] ### Integration Points - [Where new code connects to existing system] ## Specific Ideas [Any particular references, examples, or "I want it like X" moments from discussion] [If none: "No specific requirements — open to standard approaches"] ## Deferred Ideas [Ideas that came up but belong in other phases. Don't lose them.] [If none: "None — discussion stayed within phase scope"] --- *Phase: XX-name* *Context gathered: [date]* ``` Write file. Present summary and next steps: ``` Created: .planning/phases/${PADDED_PHASE}-${SLUG}/${PADDED_PHASE}-CONTEXT.md ## Decisions Captured ### [Category] - [Key decision] ### [Category] - [Key decision] [If deferred ideas exist:] ## Noted for Later - [Deferred idea] — future phase --- ## ▶ Next Up **Phase ${PHASE}: [Name]** — [Goal from ROADMAP.md] `/gsd:plan-phase ${PHASE}` <sub>`/clear` first → fresh context window</sub> --- **Also available:** - `/gsd:plan-phase ${PHASE} --skip-research` — plan without research - Review/edit CONTEXT.md before continuing --- ``` Commit phase context (uses `commit_docs` from init internally): ```bash node "$HOME/.claude/get-shit-done/bin/gsd-tools.cjs" commit "docs(${padded_phase}): capture phase context" --files "${phase_dir}/${padded_phase}-CONTEXT.md" ``` Confirm: "Committed: docs(${padded_phase}): capture phase context" Update STATE.md with session info: ```bash node "$HOME/.claude/get-shit-done/bin/gsd-tools.cjs" state record-session \ --stopped-at "Phase ${PHASE} context gathered" \ --resume-file "${phase_dir}/${padded_phase}-CONTEXT.md" ``` Commit STATE.md: ```bash node "$HOME/.claude/get-shit-done/bin/gsd-tools.cjs" commit "docs(state): record phase ${PHASE} context session" --files .planning/STATE.md ``` Check for auto-advance trigger: 1. Parse `--auto` flag from $ARGUMENTS 2. Read `workflow.auto_advance` from config: ```bash AUTO_CFG=$(node "$HOME/.claude/get-shit-done/bin/gsd-tools.cjs" config-get workflow.auto_advance 2>/dev/null || echo "false") ``` **If `--auto` flag present AND `AUTO_CFG` is not true:** Persist auto-advance to config (handles direct `--auto` usage without new-project): ```bash node "$HOME/.claude/get-shit-done/bin/gsd-tools.cjs" config-set workflow.auto_advance true ``` **If `--auto` flag present OR `AUTO_CFG` is true:** Display banner: ``` ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ GSD ► AUTO-ADVANCING TO PLAN ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ Context captured. Spawning plan-phase... ``` Spawn plan-phase as Task with direct workflow file reference (do NOT use Skill tool — Skills don't resolve inside Task subagents): ``` Task( prompt=" You are the plan-phase orchestrator. Create executable plans for Phase ${PHASE}: ${PHASE_NAME}, then auto-advance to execution. @~/.claude/get-shit-done/workflows/plan-phase.md @~/.claude/get-shit-done/references/ui-brand.md @~/.claude/get-shit-done/references/model-profile-resolution.md PHASE=${PHASE} ARGUMENTS='${PHASE} --auto' 1. Read plan-phase.md from execution_context for your complete workflow 2. Follow ALL steps: initialize, validate, load context, research, plan, verify, auto-advance 3. When spawning agents (gsd-phase-researcher, gsd-planner, gsd-plan-checker), use Task with specified subagent_type and model 4. For step 14 (auto-advance to execute): spawn execute-phase as a Task with DIRECT file reference — tell it to read execute-phase.md. Include @file refs to execute-phase.md, checkpoints.md, tdd.md, model-profile-resolution.md. Pass --no-transition flag so execute-phase returns results instead of chaining further. 5. Do NOT use the Skill tool or /gsd: commands. Read workflow .md files directly. 6. Return: PHASE COMPLETE (full pipeline success), PLANNING COMPLETE (planning done but execute failed/skipped), PLANNING INCONCLUSIVE, or GAPS FOUND ", subagent_type="general-purpose", description="Plan Phase ${PHASE}" ) ``` **Handle plan-phase return:** - **PHASE COMPLETE** → Full chain succeeded. Display: ``` ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ GSD ► PHASE ${PHASE} COMPLETE ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ Auto-advance pipeline finished: discuss → plan → execute Next: /gsd:discuss-phase ${NEXT_PHASE} --auto <sub>/clear first → fresh context window</sub> ``` - **PLANNING COMPLETE** → Planning done, execution didn't complete: ``` Auto-advance partial: Planning complete, execution did not finish. Continue: /gsd:execute-phase ${PHASE} ``` - **PLANNING INCONCLUSIVE / CHECKPOINT** → Stop chain: ``` Auto-advance stopped: Planning needs input. Continue: /gsd:plan-phase ${PHASE} ``` - **GAPS FOUND** → Stop chain: ``` Auto-advance stopped: Gaps found during execution. Continue: /gsd:plan-phase ${PHASE} --gaps ``` **If neither `--auto` nor config enabled:** Route to `confirm_creation` step (existing behavior — show manual next steps). - Phase validated against roadmap - Codebase scouted for reusable assets, patterns, and integration points - Gray areas identified through intelligent analysis with code context annotations - User selected which areas to discuss - Each selected area explored until user satisfied (with code-informed options) - Scope creep redirected to deferred ideas - CONTEXT.md captures actual decisions, not vague vision - CONTEXT.md includes code_context section with reusable assets and patterns - Deferred ideas preserved for future phases - STATE.md updated with session info - User knows next steps