Agent skill
using-superspec
Use at the start of any conversation - establishes how to find and use SuperSpec skills. Combines development discipline with spec-driven documentation.
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/using-superspec
SKILL.md
Using SuperSpec
SuperSpec combines development discipline with spec-driven documentation.
Core Principles
| Principle | Description |
|---|---|
| Spec First | All development has Spec as source of truth |
| TDD Enforced | Write test first, watch it fail, then implement |
| Two-Stage Review | Spec compliance → Code quality |
| Evidence First | Verification over claims |
| Delta Tracking | Structured change history |
| Archive Everything | Complete development documentation |
The Unified Workflow
┌─────────────────────────────────────────────────────────────────────────┐
│ FULL WORKFLOW (large features, team review needed) │
├─────────────────────────────────────────────────────────────────────────┤
│ /superspec:brainstorm → Progressive design (Explore → Propose → Spec)│
│ ↓ │
│ superspec validate → Validate specs (CLI) + team review │
│ ↓ │
│ /superspec:plan → Create TDD implementation plan │
└─────────────────────────────────────────────────────────────────────────┘
┌─────────────────────────────────────────────────────────────────────────┐
│ FAST TRACK (small-medium features, solo development) │
├─────────────────────────────────────────────────────────────────────────┤
│ /superspec:kickoff → All-in-one: brainstorm + validate + plan │
└─────────────────────────────────────────────────────────────────────────┘
↓ (both paths continue with)
/superspec:execute → Subagent-driven TDD implementation
↓
/superspec:verify → Verify implementation matches specs
↓
/superspec:finish-branch → Complete branch (merge/PR)
↓
/superspec:archive → Archive change, apply deltas
Brainstorm Phases
/superspec:brainstorm
│
├─→ Phase 1: EXPLORE
│ • Free exploration
│ • Clarifying questions
│ • Visualize ideas
│
├─→ Phase 2: PROPOSE
│ • Define Why + What Changes
│ • Define Capabilities + Impact
│ • Output: proposal.md
│
├─→ Phase 3: DESIGN
│ • 2-3 approaches comparison
│ • Technical decisions
│ • Output: design.md
│
└─→ Phase 4: SPEC
• Define Requirements
• Define Scenarios (→ tests)
• Output: specs/*.md
How to Access Skills
In Claude Code: Use the Skill tool. When you invoke a skill, its content is loaded - follow it directly.
The Rule
Invoke relevant skills BEFORE any response or action.
Even a 1% chance a skill might apply means invoke it to check.
Skill Categories
Design Phase
| Skill | When to Use |
|---|---|
kickoff |
Fast track - idea to plan in one session (small-medium features) |
brainstorm |
Full workflow - progressive design with review points (large features) |
Planning Phase
| Skill | When to Use |
|---|---|
plan-writing |
Creating TDD implementation plan |
git-worktree |
Setting up isolated development environment |
Implementation Phase
| Skill | When to Use |
|---|---|
tdd |
Writing any implementation code |
subagent-development |
Executing plan with subagents (default) |
executing-plans |
Batch execution with checkpoints (alternative) |
dispatching-parallel-agents |
For 2+ independent tasks in parallel |
systematic-debugging |
Debugging issues |
Quality & Discipline Phase
| Skill | When to Use |
|---|---|
verification-before-completion |
Before marking any task complete - evidence first |
receiving-code-review |
Responding to code review feedback professionally |
Review Phase
| Skill | When to Use |
|---|---|
spec-validation |
Validating specifications |
code-review |
Reviewing implementation |
Completion Phase
| Skill | When to Use |
|---|---|
superspec:verify |
Verifying implementation matches specs |
superspec:finish-branch |
Completing branch (merge/PR/keep/discard) |
superspec:archive |
Archiving completed changes (after finish-branch) |
Red Flags - STOP and Check Skills
| Thought | Reality |
|---|---|
| "This is just a simple question" | Questions are tasks. Check for skills. |
| "I need more context first" | Skill check comes BEFORE clarifying questions. |
| "Let me explore the codebase first" | Skills tell you HOW to explore. Check first. |
| "I'll just write this code quickly" | TDD skill is REQUIRED. No exceptions. |
| "Tests after achieve the same goals" | NO. Tests-first = "what should this do?" |
| "I know what that means" | Knowing concept ≠ using skill. Invoke it. |
| "This doesn't need a spec" | If behavior changes, it needs a spec. |
| "I'll document later" | Documentation is part of the workflow. Do it now. |
Skill Priority
When multiple skills could apply:
- Process skills first (brainstorm) - determine approach through 4 phases
- Implementation skills second (tdd, subagent-development) - guide execution
- Completion skills last (verify, archive) - finalize work
Document Outputs
| Phase | Skill | Output |
|---|---|---|
| Brainstorm | brainstorm |
proposal.md + design.md + specs/**/*.md |
| Plan | plan-writing |
superspec/changes/[id]/plan.md + tasks.md |
| Execute | subagent-development |
Actual code + tests |
| Finish | finish-branch |
Merged branch or PR created |
| Archive | archive |
superspec/changes/archive/YYYY-MM-DD-[id]/ |
Validation Checkpoints
After Spec Phase
superspec validate [change-id] --strict
- Every Requirement has at least one Scenario
- Scenario format is correct (#### Scenario:)
- MODIFIED includes full content
- REMOVED includes Reason and Migration
After Execute Phase
superspec verify [change-id]
- Every Scenario has corresponding test
- All tests pass
- No features outside of Specs
Before Archive
- All validation passes
- Final code review approved
- Tests green
The Iron Laws
TDD Law
NO PRODUCTION CODE WITHOUT A FAILING TEST FIRST
Write code before test? Delete it. Start over.
Spec Law
SPECS ARE TRUTH. CHANGES ARE PROPOSALS.
No implementation without a Spec. No Spec without a Proposal.
SuperSpec Law
EVERY SCENARIO BECOMES A TEST. EVERY TEST TRACES TO A SCENARIO.
Spec → Test → Implementation → Verification. Full traceability.
Verification Law
NO COMPLETION CLAIMS WITHOUT FRESH VERIFICATION EVIDENCE
"It should work" is not evidence. Run the test. Show the output. Then claim completion.
Quick Reference
# CLI Commands
superspec init # Initialize project
superspec list # List changes
superspec list --specs # List specs
superspec show [item] # Show details
superspec validate [id] --strict # Validate specs
superspec verify [id] # Verify implementation
superspec archive [id] --yes # Archive change
# Slash Commands (AI Assistant)
/superspec:kickoff # Fast track: brainstorm + validate + plan
/superspec:brainstorm # Full workflow: progressive design only
/superspec:plan # Create TDD plan (after brainstorm)
/superspec:execute # Execute with subagents
/superspec:verify # Verify implementation
/superspec:finish-branch # Complete branch (merge/PR)
/superspec:archive # Archive change
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.
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.
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.
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.
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.
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.
Didn't find tool you were looking for?