Skip to content

README.md

Latest commit

 

History

History
135 lines (106 loc) · 3.88 KB

README.md

File metadata and controls

135 lines (106 loc) · 3.88 KB
Raw

Creer

AI-powered repo scaffolding inside your workspace.

Describe an idea → Creer plans a clean structure → optionally preview/confirm → writes files into your VS Code workspace → optionally initializes git and creates a GitHub remote.

Current version: 0.4.0

Architecture

creer/
├── backend/     # Python FastAPI AI engine
└── extension/   # VS Code extension

Prerequisites

  • Python 3.10+
  • Node.js 18+
  • OpenAI API key or a local OpenAI-compatible server (OPENAI_BASE_URL), unless using offline template mode (CREER_OFFLINE=1)
  • VS Code / Cursor

Backend (v0.4)

cd backend
python -m venv venv
source venv/bin/activate   # Windows: venv\Scripts\activate
pip install -r requirements.txt
cp .env.example .env
# set OPENAI_API_KEY and/or OPENAI_BASE_URL in .env
uvicorn main:app --reload --port 8000

Environment

Variable Description
OPENAI_API_KEY OpenAI API key (optional if using OPENAI_BASE_URL or CREER_OFFLINE=1)
OPENAI_BASE_URL OpenAI-compatible base URL (Ollama, LM Studio, etc.). Example: http://127.0.0.1:11434/v1
CREER_MODEL Model name (default gpt-4o-mini)
CREER_OFFLINE 1 / true / yes for template-only stubs (planning requires template_id)

Endpoints

Method Path Description
GET /health Version 0.4.0, offline flag, model
GET /templates Curated starter templates
GET /bakeins License + CI bake-in options
POST /plan Plan only (no file contents)
POST /generate Generate files (+ bake-ins + quality)
POST /generate/stream SSE progress; start includes job_id
POST /generate/cancel Cancel by job_id
POST /quality Dry-run quality gates
POST /github/create-repo Create GitHub repo

Generate body (sync or stream)

{
  "idea": "Build a FastAPI todo app",
  "template_id": "fastapi-minimal",
  "plan": null,
  "job_id": null,
  "bakeins": {
    "license": "mit",
    "ci": "auto",
    "include_readme": true
  }
}

bakeins.license: mit | apache-2.0 | none
bakeins.ci: auto | python | node | none

Stream events

Event Meaning
start Includes job_id, total, project_name
file Per-file generating / done
done Full files map + quality
cancelled Stopped via /generate/cancel
error Failure detail 
curl -N -X POST http://localhost:8000/generate/stream \
  -H "Content-Type: application/json" \
  -d '{"idea":"todo api","template_id":"fastapi-minimal"}'

Extension (v0.4)

cd extension
npm install
npm run compile
  1. Open extension/ → F5
  2. Open a workspace folder
  3. Creer: Create New Repo
  4. Pick template → license/CI → confirm preview → generate (cancellable while streaming)

Commands

Command Title
creer.createRepo Creer: Create New Repo
creer.createRepoFromChat Creer: Create from Chat Prompt
creer.setGitHubToken Creer: Set GitHub Token
creer.clearGitHubToken Creer: Clear GitHub Token

Settings

Setting Default Description
creer.backendUrl http://localhost:8000 Backend base URL
creer.initGit true git init + initial commit
creer.previewBeforeWrite true Confirm plan before generate
creer.useStreaming true SSE progress (cancellable)
creer.promptBakeins true QuickPick license/CI each run
creer.license mit Used when not prompting
creer.ciPreset auto Used when not prompting
creer.createGitHubRepo false Create GitHub remote after scaffold
creer.githubPrivate true Private GitHub repos
creer.githubToken "" Deprecated — use SecretStorage

GitHub push uses GIT_ASKPASS (token never on argv/URL).

License

MIT