CLI tool for integrating Iriai platform modules into local projects
Project description
iriai-cli
Command-line tool for integrating Iriai platform modules into your local projects.
Installation
pip install iriai-cli
# or
pipx install iriai-cli
Usage
# List available modules
iriai modules list
# Analyze your project
iriai analyze
# Install a module
iriai integrate auth-react --action install
# Repair a misconfigured module
iriai integrate auth-react --action repair
# Install multiple modules
iriai integrate auth-react auth-python pwa --action install
Features
- Local file integration (no Git commits required)
- Pre-flight analysis and misconfiguration detection
- LLM-powered code generation via engine API or locally via Claude Code (
--local) - Interactive file write prompts (Apply/Skip/Overwrite-all)
- Syntax-highlighted diffs
- Secret file filtering (
.env,*.pem,*.key,credentials.*, etc.) - Dry-run mode for previewing changes
Configuration
Environment Variables
| Variable | Default | Description |
|---|---|---|
IRIAI_ENGINE_URL |
https://integration.iriai.app |
Integration engine base URL |
IRIAI_API_TOKEN |
(none) | Bearer token for authenticated modules (get from deploy console) |
IRIAI_MODULES_DIR |
(fetched from API) | Override: load module definitions from local YAML files instead of API |
# Point to local engine during development
export IRIAI_ENGINE_URL=http://localhost:8005
# Set API token for authenticated modules
export IRIAI_API_TOKEN=your_token_here
Commands
iriai modules list
List all available modules with their versions and descriptions.
iriai analyze
Analyze the current directory for module compatibility and misconfigurations. Creates a tarball of the project directory (excluding secrets and ignored files) and uploads it to the engine's POST /api/analyze-local endpoint.
iriai integrate <modules...>
Integrate one or more modules into your project.
Options:
--action:install,repair, oroverwrite(default:install)--depth:coreorfull(default:full)--yes: Skip interactive prompts, apply all changes--dry-run: Show what would be changed without writing files--local: Run entirely locally via Claude Code (claude -p) instead of the engine API
Local Mode (--local)
The --local flag runs LLM integration on your machine via Claude Code instead of the engine API:
iriai integrate auth-react --local
This mode:
- Fetches module definitions from the integration engine API
- Analyzes the project locally (same logic as the engine)
- Builds a prompt identical to
LLMIntegrator._build_prompt() - Invokes
claude -psubprocess with a 5-minute timeout - Parses the structured JSON response
- Applies file operations through interactive prompts
Requirements: Claude Code CLI (claude) must be installed and on your PATH.
Note: Profile matching is skipped in local mode (requires database). Most integrations work without profile-specific config.
For development, set IRIAI_MODULES_DIR to load module definitions from local YAML files instead of the API.
License
MIT
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file iriai_cli-0.3.0.tar.gz.
File metadata
- Download URL: iriai_cli-0.3.0.tar.gz
- Upload date:
- Size: 37.1 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.11.9
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
097371286679b83ecc736958bc115b23bdbfe00b53c926d87dd2dea5f8c8417d
|
|
| MD5 |
ed40d3eec6c979c0d6bc8a97f79658fa
|
|
| BLAKE2b-256 |
254e59de8e8f316608477855edc0a3bbd9c94f2ce22cc55023dd2c30365d72c8
|
File details
Details for the file iriai_cli-0.3.0-py3-none-any.whl.
File metadata
- Download URL: iriai_cli-0.3.0-py3-none-any.whl
- Upload date:
- Size: 40.1 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.11.9
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
18fb68e4190a8dd2b643b373840b23e9f3319d239778a68b8df2d1e75ffc06ad
|
|
| MD5 |
5b9f341b93118b3f79f76950f0fb782e
|
|
| BLAKE2b-256 |
900df67e7c1031312ce151ff621314871acda5273a50efa910d3922fb67abdcc
|