specflow-cc 1.0.0

package/commands/sf/split.md

@@ -0,0 +1,341 @@
+---
+name: sf:split
+description: Split a large specification into smaller sub-specifications
+allowed-tools:
+  - Read
+  - Write
+  - Bash
+  - Glob
+  - Grep
+  - Task
+  - AskUserQuestion
+---
+
+<purpose>
+Analyze a specification's complexity and split it into smaller, manageable sub-specifications with proper dependency chains. Large specifications (>150k tokens estimated) should be decomposed before implementation.
+</purpose>
+
+<context>
+@.specflow/STATE.md
+@.specflow/specs/SPEC-*.md
+@~/.claude/specflow-cc/agents/spec-splitter.md
+</context>
+
+<arguments>
+- `[ID]` — Specification ID to split (e.g., SPEC-001). Optional — defaults to active specification.
+</arguments>
+
+<workflow>
+
+## Step 1: Verify Initialization
+
+```bash
+[ -d .specflow ] && echo "OK" || echo "NOT_INITIALIZED"
+```
+
+**If NOT_INITIALIZED:**
+```
+SpecFlow not initialized.
+
+Run `/sf init` first.
+```
+Exit.
+
+## Step 2: Determine Target Specification
+
+**If ID provided:**
+```bash
+[ -f ".specflow/specs/{ID}.md" ] && echo "FOUND" || echo "NOT_FOUND"
+```
+
+**If NOT_FOUND:**
+```
+Specification {ID} not found.
+
+Use `/sf list` to see available specifications.
+```
+Exit.
+
+**If no ID provided:**
+Read active specification from `.specflow/STATE.md`:
+- Parse "Active Specification" field
+- Use that ID
+
+**If no active specification:**
+```
+No specification specified and no active specification.
+
+Usage: `/sf split SPEC-001`
+   or: Set active spec with `/sf show SPEC-001`
+```
+Exit.
+
+## Step 3: Check Complexity
+
+Read the specification and assess complexity:
+- Count sections and requirements
+- Estimate based on scope described
+- Check frontmatter `complexity` field
+
+**If complexity is small:**
+```
+━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
+ SPLIT NOT NEEDED
+━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
+
+**Specification:** {ID}
+**Complexity:** small (~{estimate}k tokens)
+
+This specification is small enough to implement directly.
+
+**Next Step:** `/sf audit {ID}` — audit before implementation
+```
+Exit.
+
+**If complexity is medium:**
+```
+━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
+ SPLIT OPTIONAL
+━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
+
+**Specification:** {ID}
+**Complexity:** medium (~{estimate}k tokens)
+
+This specification can be implemented as-is or split for easier management.
+
+Continue with split analysis? [y/N]
+```
+
+**If user declines:** Exit.
+
+## Step 4: Spawn Spec Splitter Agent
+
+Launch the spec-splitter subagent:
+
+```
+Task(prompt="
+<specification>
+@.specflow/specs/{ID}.md
+</specification>
+
+<project_context>
+@.specflow/PROJECT.md
+</project_context>
+
+<current_state>
+@.specflow/STATE.md
+</current_state>
+
+Analyze this specification and propose a split into smaller sub-specifications.
+Follow the spec-splitter agent instructions.
+", subagent_type="sf-spec-splitter", description="Split specification")
+```
+
+## Step 5: Present Proposal
+
+Display agent's split proposal:
+
+```
+━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
+ SPLIT ANALYSIS: {ID}
+━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
+
+**Current:** {ID} — {title}
+**Complexity:** {complexity} (~{estimate}k tokens)
+**Status:** {status}
+
+---
+
+## Analysis
+
+{Agent's analysis of structure and boundaries}
+
+## Proposed Split
+
+| ID        | Title                    | Est. Size | Dependencies |
+|-----------|--------------------------|-----------|--------------|
+| {ID}a     | {title}                  | small     | —            |
+| {ID}b     | {title}                  | medium    | {ID}a        |
+| {ID}c     | {title}                  | small     | {ID}b        |
+
+## Dependency Graph
+
+```
+{ID}a ({short description})
+    ↓
+{ID}b ({short description})
+    ↓
+{ID}c ({short description})
+```
+
+---
+
+**Options:**
+1. Accept split as proposed
+2. Modify split (describe changes)
+3. Cancel (keep as single spec)
+
+Your choice [1]:
+```
+
+## Step 6: Handle User Choice
+
+**If 1 (Accept):**
+- Agent creates child specifications
+- Agent archives parent
+- Agent updates STATE.md
+- Continue to Step 7
+
+**If 2 (Modify):**
+- Use AskUserQuestion to get modifications
+- Re-run agent with modifications
+- Return to Step 5
+
+**If 3 (Cancel):**
+```
+Split cancelled.
+
+Specification {ID} unchanged.
+```
+Exit.
+
+## Step 7: Display Result
+
+```
+━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
+ SPLIT COMPLETE
+━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
+
+**Parent:** {ID} (archived)
+**Children:** {N} specifications created
+
+### Created Specifications
+
+| ID        | Title                    | Size   | Depends On   |
+|-----------|--------------------------|--------|--------------|
+| {ID}a     | {title}                  | small  | —            |
+| {ID}b     | {title}                  | medium | {ID}a        |
+| {ID}c     | {title}                  | small  | {ID}b        |
+
+### Dependency Graph
+
+```
+{ID}a
+    ↓
+{ID}b
+    ↓
+{ID}c
+```
+
+### Files
+
+**Created:**
+- .specflow/specs/{ID}a.md
+- .specflow/specs/{ID}b.md
+- .specflow/specs/{ID}c.md
+
+**Archived:**
+- .specflow/archive/{ID}.md
+
+---
+
+## Next Step
+
+`/sf audit {ID}a` — start with first sub-specification (no dependencies)
+
+**Tip:** Use `/sf deps` to see the full dependency graph.
+```
+
+</workflow>
+
+<fallback>
+
+**If agent spawning fails**, execute inline:
+
+## Inline Split Analysis
+
+### Read Specification
+
+```bash
+cat .specflow/specs/{ID}.md
+```
+
+### Identify Boundaries
+
+Based on specification content, identify:
+1. Data/model layer tasks
+2. Logic/service layer tasks
+3. Integration/API layer tasks
+4. UI/presentation layer tasks
+
+### Generate Sub-Spec IDs
+
+Parent: SPEC-001 → Children: SPEC-001a, SPEC-001b, SPEC-001c, etc.
+
+### Create Child Specifications
+
+For each child, create `.specflow/specs/{ID}{letter}.md`:
+
+```markdown
+---
+id: {ID}{letter}
+parent: {ID}
+type: {inherited from parent}
+status: draft
+priority: {inherited from parent}
+complexity: {estimated}
+depends_on: [{previous child or empty}]
+created: {today}
+---
+
+# {Scoped Title}
+
+> Part {N} of {total} from {ID} ({parent title})
+
+## Context
+{Extracted from parent, scoped to this part}
+
+## Task
+{Scoped to this sub-spec only}
+
+## Requirements
+{Relevant subset from parent}
+
+## Acceptance Criteria
+{Relevant criteria from parent}
+
+## Constraints
+{Inherited from parent}
+
+## Assumptions
+{Inherited from parent}
+```
+
+### Archive Parent
+
+```bash
+mkdir -p .specflow/archive
+mv .specflow/specs/{ID}.md .specflow/archive/
+```
+
+Add split reference to archived parent.
+
+### Update STATE.md
+
+- Remove parent from Queue
+- Add children to Queue
+- Set first child as Active Specification
+
+</fallback>
+
+<success_criteria>
+- [ ] Initialization verified
+- [ ] Target specification identified
+- [ ] Complexity assessed
+- [ ] Split proposal generated with dependencies
+- [ ] User approved (or cancelled)
+- [ ] Child specifications created with proper frontmatter
+- [ ] Parent archived with split reference
+- [ ] STATE.md updated
+- [ ] Clear next step provided
+</success_criteria>

package/commands/sf/status.md

@@ -0,0 +1,170 @@
+---
+name: sf:status
+description: Show current SpecFlow state and recommended next step
+allowed-tools:
+  - Read
+  - Bash
+  - Glob
+---
+
+<purpose>
+Display the current SpecFlow state including active specification, queue, and recommended next action. Provides situational awareness for continuing work.
+</purpose>
+
+<context>
+@.specflow/STATE.md
+@.specflow/PROJECT.md
+</context>
+
+<workflow>
+
+## Step 1: Verify Initialization
+
+```bash
+[ -d .specflow ] && echo "OK" || echo "NOT_INITIALIZED"
+```
+
+**If NOT_INITIALIZED:**
+```
+SpecFlow not initialized.
+
+Run `/sf init` to start.
+```
+Exit.
+
+## Step 2: Load State
+
+Read `.specflow/STATE.md` and extract:
+- Active Specification
+- Status
+- Next Step
+- Queue
+- Recent Decisions
+- Warnings
+
+## Step 3: Load Project Info
+
+Read `.specflow/PROJECT.md` and extract:
+- Project name
+- Tech stack summary
+
+## Step 4: Count Statistics
+
+```bash
+# Total specs
+TOTAL=$(ls .specflow/specs/SPEC-*.md 2>/dev/null | wc -l)
+
+# Completed specs
+DONE=$(ls .specflow/archive/SPEC-*.md 2>/dev/null | wc -l)
+
+# Pending todos
+TODOS=$(ls .specflow/todos/*.md 2>/dev/null | wc -l)
+```
+
+## Step 5: Determine Next Action
+
+Based on current status:
+
+| Status | Next Action |
+|--------|-------------|
+| idle | `/sf new "task"` — create specification |
+| drafting | `/sf audit` — audit specification |
+| revision_requested | `/sf revise` — fix issues |
+| audited | `/sf run` — implement |
+| running | Continue implementation or `/sf review` |
+| reviewing | `/sf fix` or `/sf done` |
+
+## Step 6: Display Status
+
+```
+━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
+ SPECFLOW STATUS
+━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
+
+**Project:** [name]
+**Progress:** [████████░░] {done}/{total} specs complete
+
+---
+
+## Current Position
+
+**Active:** {SPEC-XXX: Title | none}
+**Status:** {status}
+
+{If active spec exists:}
+**Type:** {feature | refactor | bugfix}
+**Complexity:** {small | medium | large}
+**Priority:** {high | medium | low}
+
+---
+
+## Queue
+
+| # | ID | Title | Priority | Status |
+|---|----|-------|----------|--------|
+{Queue from STATE.md}
+
+{If queue empty:}
+No specifications in queue.
+
+---
+
+{If recent decisions exist:}
+## Recent Decisions
+
+| Spec | Decision |
+|------|----------|
+{Last 3-5 decisions}
+
+---
+
+{If warnings exist:}
+## Warnings
+
+{List warnings}
+
+---
+
+## Next Step
+
+`{recommended command}` — {description}
+
+{Additional context-specific suggestions}
+```
+
+## Step 7: Context-Specific Suggestions
+
+Based on state, provide additional guidance:
+
+**If idle for long time:**
+```
+**Tip:** Have ideas? Use `/sf todo "idea"` to capture for later.
+```
+
+**If spec is large:**
+```
+**Tip:** Consider `/sf split SPEC-XXX` to break into smaller specs.
+```
+
+**If multiple specs in queue:**
+```
+**Tip:** Use `/sf next` to work on highest priority item.
+```
+
+**If warnings present:**
+```
+**Note:** {N} warnings recorded. Review before continuing.
+```
+
+</workflow>
+
+<success_criteria>
+- [ ] Initialization verified
+- [ ] STATE.md loaded
+- [ ] PROJECT.md info extracted
+- [ ] Statistics calculated
+- [ ] Current position displayed
+- [ ] Queue shown
+- [ ] Recommended next step clear
+- [ ] Context-specific tips provided
+</success_criteria>

package/commands/sf/todo.md

@@ -0,0 +1,130 @@
+---
+name: sf:todo
+description: Add a new to-do item for future work
+allowed-tools:
+  - Read
+  - Write
+  - Bash
+  - AskUserQuestion
+---
+
+<purpose>
+Add a new to-do item to the backlog. To-dos are ideas or tasks that don't need immediate specification but should be captured for later. They can later be converted to specifications with `/sf plan`.
+</purpose>
+
+<context>
+@.specflow/todos/TODO.md
+</context>
+
+<arguments>
+- `[text]` — Optional. Description of the to-do item. If omitted, will prompt interactively.
+</arguments>
+
+<workflow>
+
+## Step 1: Verify Initialization
+
+```bash
+[ -d .specflow ] && echo "OK" || echo "NOT_INITIALIZED"
+```
+
+**If NOT_INITIALIZED:**
+```
+SpecFlow not initialized.
+
+Run `/sf init` to start.
+```
+Exit.
+
+## Step 2: Ensure Todos Directory Exists
+
+```bash
+mkdir -p .specflow/todos
+```
+
+## Step 3: Get Description
+
+**If argument provided:**
+Use provided text as description.
+
+**If no argument:**
+Use AskUserQuestion:
+- header: "Todo"
+- question: "What idea or task do you want to capture?"
+- options: (freeform text input)
+
+## Step 4: Generate TODO ID
+
+Check existing TODO.md for highest ID:
+
+```bash
+grep -oP 'TODO-\K\d+' .specflow/todos/TODO.md 2>/dev/null | sort -n | tail -1
+```
+
+If no existing todos, start with 001.
+Next ID = last + 1, zero-padded to 3 digits.
+
+## Step 5: Create or Update TODO.md
+
+**If TODO.md doesn't exist:**
+Create from template:
+
+```markdown
+# To-Do List
+
+---
+
+## TODO-{XXX} — {YYYY-MM-DD}
+**Description:** {description}
+**Priority:** —
+**Notes:** —
+
+---
+*Last updated: {YYYY-MM-DD}*
+```
+
+**If TODO.md exists:**
+Insert new todo after `# To-Do List` line:
+
+```markdown
+## TODO-{XXX} — {YYYY-MM-DD}
+**Description:** {description}
+**Priority:** —
+**Notes:** —
+
+---
+```
+
+Also update `*Last updated:` line at the bottom.
+
+## Step 6: Display Confirmation
+
+```
+━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
+ TODO CREATED
+━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
+
+**ID:** TODO-{XXX}
+**Description:** {description}
+**Priority:** — (not set)
+
+---
+
+**Actions:**
+- `/sf todos` — view all to-do items
+- `/sf plan TODO-{XXX}` — convert to specification
+- `/sf priority` — set priorities
+```
+
+</workflow>
+
+<success_criteria>
+- [ ] Initialization verified
+- [ ] Todos directory exists
+- [ ] Description obtained (from arg or prompt)
+- [ ] Unique TODO-XXX ID generated
+- [ ] TODO.md created or updated
+- [ ] New todo added at top of list
+- [ ] Last updated timestamp refreshed
+- [ ] Clear confirmation and next actions shown
+</success_criteria>