Skip to main content

Multi-agent reasoning system with specialized thinker and grader components

Project description

Agentic Fleet

A powerful fleet of AI agents for complex reasoning and task execution, combining Tree of Thoughts reasoning with advanced agent orchestration capabilities.

Features

  • Tree of Thoughts Reasoning: Multi-agent reasoning system with beam search
  • CaptainAgent: Advanced agent orchestration based on the paper "CaptainAgent: Building Reliable Autonomous Agents through Iterative Prompting"
  • Agent Fleet: Extensible collection of specialized agents for different tasks
  • Tool Integration: Rich set of tools and capabilities for agents
  • Modern Development: Built with modern Python tools (uv, PDM)

Installation

pip install agentic-fleet

Quick Start

from agentic_fleet import CaptainAgent, ReasoningAgent
from agentic_fleet.tools import CodeAnalysisTool, DataProcessingTool

# Initialize CaptainAgent with specialized tools
captain = CaptainAgent(
    name="project_captain",
    tools=[CodeAnalysisTool(), DataProcessingTool()],
    max_iterations=5
)

# Execute a complex task
result = captain.execute_task(
    "Analyze this codebase and suggest improvements",
    context={"repo_path": "./my_project"}
)

# Use reasoning agent for complex problem-solving
reasoner = ReasoningAgent(
    name="math_reasoner",
    beam_size=3,
    max_depth=5
)

solution = reasoner.solve("What is the optimal strategy for...")

Architecture

The package consists of several key components:

  1. CaptainAgent: Orchestrates complex tasks through iterative prompting
  2. ReasoningAgent: Implements Tree of Thoughts for complex reasoning
  3. SpecialistAgents: Task-specific agents (Coder, Analyst, etc.)
  4. Tools: Extensible set of capabilities for agents
  5. Prompts: Carefully crafted prompts for different scenarios

Development

# Install development dependencies
uv venv
source .venv/bin/activate
uv pip install -e ".[dev]"

# Run tests
pytest

# Run linting
ruff check .
black .
mypy .

Contributing

See CONTRIBUTING.md for guidelines.

License

MIT License - see LICENSE for details.

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

agentic_fleet-0.1.5.tar.gz (34.7 kB view details)

Uploaded Source

Built Distribution

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

agentic_fleet-0.1.5-py3-none-any.whl (48.5 kB view details)

Uploaded Python 3

File details

Details for the file agentic_fleet-0.1.5.tar.gz.

File metadata

  • Download URL: agentic_fleet-0.1.5.tar.gz
  • Upload date:
  • Size: 34.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.0.1 CPython/3.11.5

File hashes

Hashes for agentic_fleet-0.1.5.tar.gz
Algorithm Hash digest
SHA256 3f7ee35ae6f05f4b94de9d4e588742834f71a29e954eed1446e6ffc086479c37
MD5 d09357d5d6793627f2f7c41a1231784e
BLAKE2b-256 e894465270c5a1a6290ffa44a3beb4b0c3cce05eae3f1522f3d338ae55de665e

See more details on using hashes here.

File details

Details for the file agentic_fleet-0.1.5-py3-none-any.whl.

File metadata

  • Download URL: agentic_fleet-0.1.5-py3-none-any.whl
  • Upload date:
  • Size: 48.5 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.0.1 CPython/3.11.5

File hashes

Hashes for agentic_fleet-0.1.5-py3-none-any.whl
Algorithm Hash digest
SHA256 5b53e64bb94898f9cf7dab676a4e0290cdf9020e73a2461fc3c80e26508d6a9d
MD5 ba2b67374e910e761cd0fa7191913568
BLAKE2b-256 11348565e6490aff511f37dc12c9ae3fa79c8271b6568ab62809dfcaaf27ea34

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