Menu
Show an overview of this template — what it is, available commands, and how to start.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
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.
Based on SOC occupation classification
| name | help |
| description | Show an overview of this template — what it is, available commands, and how to start. |
Output the following text verbatim:
Production-ready Python SaaS backend. Stack: Python 3.12+ · Litestar · SQLAlchemy 2.0 async · Dishka DI · msgspec · PostgreSQL · Redis · S3 · Alembic · Granian.
/specialize — Claude will ask about your domain, auth needs, and MVP scope, then configure the project for youSetup
/specialize — one-time project configuration: renames the package, prunes unused code, scaffolds your entities and endpoints, generates the initial migrationImplementation
/impl [small|mid|large] <change> — implement anything; Claude auto-detects the right pipeline (small = in-place edit, mid = chain of /add-* skills, large = full planning cycle); pass a tier to override/chat <problem> — design consultation before committing to an approach; Claude researches patterns and presents options with trade-offs; no code writtenPlanning
/plan <feature> — produce a detailed implementation plan; used inside the large pipeline/research <topic> — investigate the codebase before making changesBuilding blocks (used inside /impl mid, or directly)
/add-endpoint <domain> <action> <method> — full endpoint pipeline/add-entity <name> <fields> — domain entity + migration/add-repository <entity> — protocol + impl + gateway wire/add-handler <domain> <action> — command + handler/add-port <category> <name> — protocol + adapter + DI wire/add-event <name> — event + publisher + handler/add-http-client <service> — HTTP client + adapter/add-migration <description> — Alembic autogenerate/add-service <domain> <name> — application service/add-test <domain> <action> — integration testTesting
/run-tests <domain | path | category> — scoped test run, faster than just test/debug-failing-test <test-path> — diagnose a failing test with project-specific recipesKnowledge
/update-knowledge [topic] — update .claude/ docs to reflect current codebase state/explain <question> — quick answer from .claude/ docs; no codebase scanjust check ruff format + ruff check + mypy + layer guard
just run api start the API server
just run queue start the job queue worker
just test integration tests (requires Docker)
just test-unit unit tests (no Docker)
just migrate apply pending migrations
just migration "x" autogenerate a new migration
just start start Docker Compose stack
just stop stop Docker Compose stack