القائمة
Answer a quick question about the template using only .claude/ docs — no codebase scan
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
| name | explain |
| description | Answer a quick question about the template using only .claude/ docs — no codebase scan |
| argument-hint | <question> |
| disable-model-invocation | true |
Answer a quick question about why the template is structured the way it is. Answers come only from the .claude/ docs — no codebase scan, no file reading beyond the docs. Fast and low-effort by design.
This skill is for simple conceptual questions. If you need to trace actual code, understand a specific flow in detail, or investigate something that may have changed since the docs were last updated, use /research <topic> instead.
$ARGUMENTS — the question (e.g. "why is there a domain and app layer", "what is a port", "how does DI work")If $ARGUMENTS is empty, print:
Usage: /explain <question>
Examples:
/explain why there is a domain and app layer
/explain what is a port and why not just use a class
/explain how does dependency injection work
/explain when should I use a service vs a handler
For deep code tracing, use /research <topic> instead.
Then stop.
Check the question for keywords and select the 1–3 most relevant rule files to read. Do not read all rules — pick only what matches.
| Keywords in question | Rule file |
|---|---|
| layer, import direction, domain, app, entry, infra, internal, why separate | architecture.md |
| entity, table, ORM, mixin, column, mapped, relationship | entities.md |
| handler, command, use case, business logic, write operation | handlers.md |
| repository, repo, gateway, CRUDSupported, ImplRepoGateway | repositories.md |
| port, adapter, protocol, abstraction, dependency inversion | ports-adapters.md |
| DI, dependency injection, provider, Dishka, scope, wire | dependency-injection.md |
| error, exception, Result, Option, DetailedError, some | error-handling.md |
| DTO, msgspec, Struct, response shape, request body | dtos.md |
| controller, route, endpoint, HTTP, decorator, inject | controllers.md |
| event, publish, queue, async, background, DBus | events.md |
| HTTP client, external API, external service, integration | external-services.md |
| test, integration test, unit test, fixture, factory | testing.md |
| type, mypy, generic, typing, annotation | typing.md |
| migration, Alembic, schema, database, column change | database.md |
| style, ruff, format, lint, comment | code-style.md |
| git, commit, branch, PR, workflow | git.md |
If no keyword clearly matches, read only CLAUDE.md (the Key Patterns and Architecture sections).
Rule files live at: .claude/rules/<file>
List what's available first if uncertain:
! ls .claude/rules/
Read the selected files. Also read the Architecture and Key Patterns sections of .claude/CLAUDE.md if the question is architectural.
Answer the question:
End every answer with this line, verbatim:
This answer is drawn from
.claude/docs only. For questions that require tracing actual code, run/research <topic>.
Add a complete REST endpoint from command to controller route. Use when building new API endpoints.
Summarize the API interface changes made during THIS session for the frontend, then optionally spawn an agent to implement them in the frontend repo. Use after changing endpoints, request/response DTOs, or auth.
One-shot guided session — prunes unused code, scaffolds new entities/endpoints, updates project identity config, and replaces the init migration for a specific project.
Compress natural language memory files (CLAUDE.md, todos, preferences) into caveman format to save input tokens. Preserves all technical substance, code, URLs, and structure. Compressed version overwrites the original file. Human-readable backup saved as FILE.original.md. Trigger: /caveman-compress FILEPATH or "compress memory file"
Implement a change with a pipeline scaled to its complexity. Single entry point for all code changes — auto-detects effort or accepts small|mid|tuff override.
Run a Linear task end-to-end — creates an isolated tmux session + worktree, starts a Claude agent to implement it, opens a PR, and watches for review feedback automatically.