Use when starting any reverse engineering task - establishes how to find and use Unwind skills for codebase analysis, service mapping, and documentation
Initiates reverse engineering workflows to map architectures and analyze codebases layer by layer.
/plugin marketplace add cliftonc/unwind/plugin install cliftonc-unwind@cliftonc/unwindThis skill inherits all available tools. When active, it can use any tool Claude has access to.
Unwind is a skills library focused on reverse engineering existing services and codebases. It provides structured approaches for understanding unfamiliar code, mapping service architectures, discovering APIs, and generating documentation.
discovering-architecture ← Start here: explore codebase
│
▼
architecture.md ← Machine-parseable layer map
│
▼
unwinding-codebase ← Orchestrates layer specialists
│
├──► analyzing-database-layer
├──► analyzing-domain-model
├──► analyzing-service-layer
├──► analyzing-api-layer
├──► analyzing-messaging-layer (if present)
└──► analyzing-frontend-layer (if present)
│
▼
layers/*.md ← Detailed layer documentation
│
▼
synthesizing-findings ← Combine into final doc
│
▼
CODEBASE.md ← Complete documentation
Start with discovery:
Use unwind:discovering-architecture
Review the architecture document:
Check docs/unwind/architecture.md for accuracy
Run layer analysis:
Use unwind:unwinding-codebase
Synthesize findings:
Use unwind:synthesizing-findings
| Skill | Purpose |
|---|---|
discovering-architecture | Initial exploration, identifies layers and patterns |
unwinding-codebase | Orchestrates layer-by-layer analysis with subagents |
synthesizing-findings | Aggregates layer docs into unified CODEBASE.md |
| Skill | Purpose |
|---|---|
analyzing-database-layer | Schema, migrations, ORM, data access patterns |
analyzing-domain-model | Entities, value objects, business rules, aggregates |
analyzing-service-layer | Business logic, transformation, DTOs, integrations |
analyzing-api-layer | REST/GraphQL endpoints, auth, contracts |
analyzing-messaging-layer | Events, queues, async patterns |
analyzing-frontend-layer | Components, state, routing, API integration |
All documentation is written to the target project:
docs/unwind/
├── architecture.md # Discovery output
├── layers/
│ ├── database.md
│ ├── domain-model.md
│ ├── service-layer.md
│ ├── api.md
│ ├── messaging.md # If messaging exists
│ └── frontend.md # If frontend exists
└── CODEBASE.md # Final synthesized documentation
All skills support incremental review:
## Changes Since Last ReviewInvoke relevant skills BEFORE diving into code. Even a 1% chance a skill might apply means you should check it.
These thoughts mean STOP - check for applicable skills:
| Thought | Reality |
|---|---|
| "Let me just grep around" | Skills provide systematic approaches |
| "I'll figure it out as I go" | Structure prevents missed components |
| "This codebase is simple" | Simple-looking code often hides complexity |
| "I'll document later" | Document as you discover |
To add new skills:
skills/ with your skill nameSKILL.md with frontmatter (name, description)superpowers:writing-skillsThis skill should be used when the user asks about libraries, frameworks, API references, or needs code examples. Activates for setup questions, code generation involving libraries, or mentions of specific frameworks like React, Vue, Next.js, Prisma, Supabase, etc.