Agent skill

sdd-orchestrator

Use when the user is starting a project or wants to catch up on project progress - scans the specs/ folder, surfaces current status, proposes the next SDD phase, and routes to the appropriate workflow after confirmation.

View SKILL.md on GitHub Repository
Stars 163
Forks 31

Install this agent skill to your Project

npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/productivity/sdd-orchestrator

SKILL.md

Spec-Driven Development Orchestrator

What It Does

  1. Scans specs/ folder structure
  2. Detects state (product docs? active specs? current phase?)
  3. Announces status clearly with visual indicators
  4. Proposes logical next step with reasoning
  5. Waits for your confirmation
  6. Routes to appropriate workflow skill

Core principle: Never proceed without confirmation. Always present options.

The Process

Step 1: Scan Project State

bash
# Check product documentation
[ -f "specs/product/mission.md" ] && \
[ -f "specs/product/roadmap.md" ] && \
[ -f "specs/product/tech-stack.md" ]
PRODUCT_COMPLETE=$?

# Find spec folders (dated format: YYYY-MM-DD-name) (limit to 5 most recent)
find specs/features -maxdepth 1 -type d -name "20*" 2>/dev/null | sort -r | head -5

Step 2: Detect Each Spec's Phase

For each spec folder:

bash
SPEC="specs/features/2025-11-09-feature-name"

# Check phase progression
if [ ! -f "$SPEC/planning/initialization.md" ]; then
  STATUS="NOT_STARTED"
elif [ ! -f "$SPEC/planning/requirements.md" ]; then
  STATUS="IN_REQUIREMENTS"
elif [ ! -f "$SPEC/spec.md" ]; then
  STATUS="IN_SPEC_WRITING"
elif [ ! -f "$SPEC/tasks.md" ]; then
  STATUS="IN_TASKS_PLANNING"
elif [ ! -f "$SPEC/verification/spec-verification.md" ]; then
  STATUS="IN_SPEC_VERIFICATION"
elif grep -q "Status.*Failed" "$SPEC/verification/spec-verification.md"; then
  STATUS="SPEC_FAILED_VERIFICATION"
elif [ ! -f "$SPEC/verification/final-verification.md" ]; then
  if grep -q "^- \[ \]" "$SPEC/tasks.md"; then
    STATUS="IN_IMPLEMENTATION"
  else
    STATUS="IMPLEMENTATION_READY"
  fi
else
  STATUS="FULLY_COMPLETE"
fi

Step 3: Announce Status

📋 Spec-Driven Development Status

Product Documentation:
[If complete]
  ✅ specs/product/mission.md
  ✅ specs/product/roadmap.md
  ✅ specs/product/tech-stack.md
[If missing]
  ⚠️  Product documentation needed

Active Specs:
[For each spec]
  [Icon] 2025-11-09-feature-name
    └─ Phase: [Phase description]
    [If implementing]
    └─ Tasks: [X/Y] complete

Next Roadmap Item:
  [ ] Feature name - Description

Status Icons:

  • ⚪ NOT_STARTED
  • 🔵 IN_REQUIREMENTS
  • 🟡 IN_SPEC_WRITING
  • 🟠 IN_TASKS_PLANNING
  • 🟣 IN_SPEC_VERIFICATION
  • ❌ SPEC_FAILED_VERIFICATION
  • 🟢 IMPLEMENTATION_READY
  • 🔄 IN_IMPLEMENTATION
  • ✅ FULLY_COMPLETE

Step 4: Propose Next Step

Based on state:

No product docs:

Proposed: Create product documentation
Required before creating specs.

Incomplete spec:

Proposed: Continue [spec-name] - [phase]
[What this phase does]

Spec ready:

Proposed: Implement [spec-name]
[X] task groups ready to build

No active specs:

Proposed: Start next roadmap item - [feature]
Or create new spec for different feature.

Step 5: Present Options

What would you like to do?

1. [Proposed action]
2. Start a new spec
3. Review completed work
4. Something else (describe)

Choose option or tell me what you'd like to work on.

STOP. Wait for response.

Step 6: Route to Workflow

State Route To
No product docs product-planning
In requirements spec-creation-workflow (Phase 2)
In spec writing spec-creation-workflow (Phase 3)
In tasks planning spec-creation-workflow (Phase 4)
In verification spec-creation-workflow (Phase 5)
Ready for implementation spec-implementation-workflow
In implementation spec-implementation-workflow (continue)

State Detection Reference

File-based detection only - no hidden state:

specs/features/YYYY-MM-DD-name/
├─ planning/initialization.md     [EXISTS = Phase 1 complete]
├─ planning/requirements.md        [EXISTS = Phase 2 complete]
├─ spec.md                         [EXISTS = Phase 3 complete]
├─ tasks.md                        [EXISTS = Phase 4 complete]
├─ verification/spec-verification.md [EXISTS = Phase 5 complete]
└─ verification/final-verification.md [EXISTS = Fully complete]

tasks.md checkboxes:
- [ ] = incomplete
- [x] = complete

roadmap.md checkboxes:
1. [ ] = not done
2. [x] = done

Red Flags

Never:

  • Proceed without user confirmation
  • Assume what user wants
  • Skip status announcement
  • Auto-route without offering alternatives

Always:

  • Scan filesystem for current state
  • Present clear status
  • Propose with reasoning
  • Offer alternatives
  • Wait for explicit choice

Integration

Routes to:

  • product-planning - First-time setup
  • spec-creation-workflow - Creating/completing specs
  • spec-implementation-workflow - Implementing specs

Returns from:

  • All workflows return here when complete
  • Orchestrator re-scans and shows updated status

Context Preservation

When returning from other workflows, briefly summarize what was accomplished to maintain continuity.

Recommended Agent Skills

Expand your agent's capabilities with these related and highly-rated skills.

majiayu000/claude-skill-registry

agent-ops-spec

Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.

163 31
Explore
majiayu000/claude-skill-registry

agent-ops-state

Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.

163 31
Explore
majiayu000/claude-skill-registry

agent-ops-spec

Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.

163 31
Explore
majiayu000/claude-skill-registry

agent-ops-testing

Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.

163 31
Explore
majiayu000/claude-skill-registry

agent-ops-testing

Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.

163 31
Explore
majiayu000/claude-skill-registry

agent-ops-state

Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.

163 31
Explore

Didn't find tool you were looking for?

Be as detailed as possible for better results