Skip to main content

Angel Claw personal AI agent framework

Project description

Angel Claw

A personal AI agent framework based on OpenClaw, using angel-recall for memory and litellm for LLM interactions.

Features

  • FastAPI Gateway: HTTP interface for interacting with the agent.
  • Multi-Channel Bridge: Connect and pair your agent with Telegram for mobile access.
  • Agent-Native Memory: Powered by angel-recall, providing long-term, evolvable memory for each session.
  • Proactive Tasks: Built-in background worker for at, every, and cron schedules.
  • Multi-Model Support: Uses litellm to connect to various providers (OpenAI, Anthropic, Ollama, etc.).

Setup

  1. Install dependencies:

    pip install angel-claw
    
  2. Configure environment: Copy .env.example to .env and fill in your API keys.

    • MODEL_KEY: Your LLM API key.
    • TELEGRAM_TOKEN: (Optional) Your bot token from @BotFather.
    cp .env.example .env
    
  3. Run the gateway:

    angel-claw
    

Usage

For detailed instructions on how to use Angel Claw and its skills system, see the Documentation.

Out-of-the-Box Capabilities

Angel Claw comes pre-loaded with several powerful features. Here are some examples of what you can do:

1. Long-Term Memory

The agent automatically stores and retrieves facts about you.

  • Prompt: "Remember that I'm allergic to peanuts."
  • Prompt (Later): "What should I avoid at the Thai restaurant?"
  • Result: Angel Claw will retrieve your allergy fact and advise you accordingly.

2. Proactive Messaging & Scheduling

Schedule tasks or reminders that the agent will trigger on its own.

  • Prompt: "Remind me to check the oven in 20 minutes."
  • Prompt: "Every day at 9 AM, check the weather in London and message me if it's raining."
  • Prompt: "Schedule a task named 'daily-report' to run 0 18 * * * that summarizes my day."

3. Web Search & Browser Automation

The agent can browse the web to find information or interact with sites.

  • Prompt: "Search for the latest news about SpaceX and summarize it."
  • Prompt: "Go to https://news.ycombinator.com and tell me the top story."

4. Custom Skill Generation

If Angel Claw doesn't have a tool, it can write one for itself.

  • Prompt: "Create a skill called 'currency_converter' that uses an API to convert USD to EUR."

Telegram Bridge

You can interact with Angel Claw from your phone using Telegram:

  1. Get a Bot Token: Create a bot using @BotFather and add TELEGRAM_TOKEN to your .env.
  2. Start your bot on Telegram and press "Start".
  3. Pair your chat: Use the /pair command with your session ID (e.g., /pair cli-default).
  4. Angel Claw will now respond to your Telegram messages using that session's memory and skills.

Basic Chat

Send a POST request to /chat:

curl -X POST http://localhost:8000/chat \
     -H "Content-Type: application/json" \
     -d '{"session_id": "user-123", "message": "Hi, I am Alex. Remember that I like Python.", "user_id": "alex"}'

Testing

pytest

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

angel_claw-0.3.1.tar.gz (22.6 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

angel_claw-0.3.1-py3-none-any.whl (23.8 kB view details)

Uploaded Python 3

File details

Details for the file angel_claw-0.3.1.tar.gz.

File metadata

  • Download URL: angel_claw-0.3.1.tar.gz
  • Upload date:
  • Size: 22.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.10.0 {"installer":{"name":"uv","version":"0.10.0","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Linux Mint","version":"22.1","id":"xia","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

File hashes

Hashes for angel_claw-0.3.1.tar.gz
Algorithm Hash digest
SHA256 007fd695cd4f561092adbe1e73b40db177d1fe3eb303f2f3d4436107cacb5bca
MD5 14aa9ee067c31be25c39f01bff1d860e
BLAKE2b-256 8baab33140038c96d7e2979ae20d43ca920ea24f76e5be69bc862674df8e5f75

See more details on using hashes here.

File details

Details for the file angel_claw-0.3.1-py3-none-any.whl.

File metadata

  • Download URL: angel_claw-0.3.1-py3-none-any.whl
  • Upload date:
  • Size: 23.8 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.10.0 {"installer":{"name":"uv","version":"0.10.0","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Linux Mint","version":"22.1","id":"xia","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

File hashes

Hashes for angel_claw-0.3.1-py3-none-any.whl
Algorithm Hash digest
SHA256 bfb6d9cffeb21ffeeeecaa43830b4e150b34f91b4cc64c988afb2574ed9a7dd5
MD5 0b0913d763bd837313775d5e485931db
BLAKE2b-256 82720ec1c91249a3c471139888b85bb67dc115a7173e836252c8c2bf6fe2fef5

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page