archi-vide
Skill1 starsChanged 4 months agoMITView on GitHub
Generate an empty, strongly-typed code architecture (no implementation), with minimal comments that document behavior and relationships between modules.
In this repository6 other files it gives its agents.
- agents-md-syncSkill
- fix-rootSkill
- micro-optimizSkill
- naming-conventionSkill
- plan-featSkill
- plan-updateSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 7 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈7
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 9
Empty: no business logic, no side effects.
- Distinctive: line 29
No real behavior, no algorithm, no persistence, no network calls.
- Distinctive: line 30
No heavy abstraction, no frameworks unless the user explicitly asks for one.
- Distinctive: line 42
Default folder layout (omit folders you don’t need):
- Distinctive: line 55
Use-case coordination: patterns/service.md
- Distinctive: line 61
Ask for the target language/runtime if unknown (TypeScript/Node, Python, Go, etc.).
- Distinctive: line 62
Ask for the primary module boundary (feature name) and the IO needs (DB?
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Empty Architecture (Archi Vide)
Generate code scaffolding that is:
- Empty: no business logic, no side effects.
- Predictable: explicit dependencies and module boundaries.
- Readable: relationships are clear from type signatures + minimal doc comments.
- Strictly typed: type composition + generics by default.
Trigger this skill whenever the user asks for:
- A project/module skeleton, scaffolding, or “architecture”
- Stubbed code, interfaces, ports/adapters, clean architecture, DDD boundaries
- “Empty implementation” files to be filled later
Output contract
Always deliver:
- File tree (final structure).
- Code for each file (or only the files the user requests).
- One short rationale explaining boundaries + dependency direction.
- An architecture note: an
archi-*.mddocument (global → details) including: why, external impacts, expected results.
Non-goals
- No real behavior, no algorithm, no persistence, no network calls.
- No heavy abstraction, no frameworks unless the user explicitly asks for one.
- No speculative features.
Required loading (selective)
Always apply:
- Naming (mandatory): Use Skill
naming-conventionfor every single identifier (files, folders, modules, namespaces, classes, types, interfaces, functions, methods, parameters, return types, variables, properties, DB tables/columns, routes).- Fallback if
naming-conventionis unavailable: camelCase functions/variables, PascalCase types/classes/components, kebab-case JS/TS/CSS files, snake_case Python/Ruby/Dart/Go files, SCREAMING_SNAKE_CASE constants — and keep namesprefix + entity (+ suffix)shaped.
- Fallback if
- Types: load references/type-rules.md.
- Stubs: load references/stubs.md.
- Architecture note: load references/architecture-doc.md.
Default folder layout (omit folders you don’t need):
domain/,application/,ports/,infra/,api/
Load exactly one language profile:
- TypeScript: languages/typescript.md
- Python: languages/python.md
- Ruby: languages/ruby.md
- Dart/Flutter: languages/dart-flutter.md
- Go: languages/go.md
Load patterns only when needed:
- IO boundaries: patterns/ports-and-adapters.md
- Persistence: patterns/repository.md
- Use-case coordination: patterns/service.md
- DTO translation: patterns/adapter.md
- Complex construction: patterns/factory.md
Generation procedure
- Ask for the target language/runtime if unknown (TypeScript/Node, Python, Go, etc.).
- Ask for the primary module boundary (feature name) and the IO needs (DB? HTTP? queue?).
2b. If
codebase-memory-mcpis available and this scaffold lands inside an existing repo, checkindex_statusfirst — index it (index_repository) if it isn't yet, then callget_architecture(aspects). The new module's folder layout and dependency direction must match the codebase's actual boundaries, not just the default layout below. No MCP, greenfield repo, or not worth indexing for a single scaffold → use the default layout as-is.
MCP names are bare, not callable as written. Prefix cbm_ under pi,
mcp__codebase-memory-mcp__ under Claude Code. Under pi, trace_path,
query_graph, detect_changes, get_architecture and get_graph_schema
need one cbm_search_tools call first.
3. Load only the files required by the context (language + patterns).
4. Produce a file tree including archi-<feature>.md.
5. Generate strictly-typed stubs that compile and remain obviously empty.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

