Skip to main content

Zero-code AI agent monitoring with Human-in-the-Loop for LangChain, LangGraph, CrewAI, AutoGen, and any agentic framework

Project description

HumanLayer

Zero-code AI agent monitoring with Human-in-the-Loop (HITL) for any agentic framework.

PyPI version Python 3.10+ License: MIT

Install

pip install humanlayer

Quick Start — 3 lines

import humanlayer

humanlayer.init(
    api_key="adr_your_key_here",
    project_id="my-agent",
)

# That's it. All LangChain/LangGraph events are now captured automatically.

Human-in-the-Loop

tools = humanlayer.wrap_tools(
    tools,
    approval_required=[
        "send_email",
        "book_meeting",
        "create_github_issue",
    ]
)

# Agent pauses before these tools.
# Approve or reject from the HumanLayer dashboard.

Works with any agentic framework

  • LangChain — auto-instrumented via callback handler
  • LangGraph — persistent session tracking
  • CrewAI — wrap tools with HITL gates
  • AutoGen — intercept agent actions
  • Any framework using Python functions as tools

Dashboard

After calling humanlayer.init(), visit your dashboard to:

  • See every LLM call, tool use, and agent decision
  • Approve or reject tool calls before they execute
  • Review the full agent journey for each HITL decision point
  • Manage API keys, projects, and team members

Backend

Health check: GET https://hitl-agent-v1.preview.emergentagent.com/api/health

Environment Variables

Variable Description Default
HUMANLAYER_API_KEY Your API key (adr_...)
HUMANLAYER_PROJECT_ID Project identifier
HUMANLAYER_API_BASE_URL Backend URL https://hitl-agent-v1.preview.emergentagent.com
HUMANLAYER_DEBUG Enable debug logging false

License

MIT

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

humanlayer_ai-0.1.0.tar.gz (12.1 kB view details)

Uploaded Source

Built Distribution

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

humanlayer_ai-0.1.0-py3-none-any.whl (13.0 kB view details)

Uploaded Python 3

File details

Details for the file humanlayer_ai-0.1.0.tar.gz.

File metadata

  • Download URL: humanlayer_ai-0.1.0.tar.gz
  • Upload date:
  • Size: 12.1 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.14

File hashes

Hashes for humanlayer_ai-0.1.0.tar.gz
Algorithm Hash digest
SHA256 2d4cf98bf6ef97f9f027584c7c9e759e5eaac958f5807019ae65c2c666c2e7b1
MD5 6ed979b17c6df15f32e26fdacc5ea05a
BLAKE2b-256 7de41d85b24d998ed353f915c4606f7e7ee66edcf0fcaf84815490853993dad4

See more details on using hashes here.

File details

Details for the file humanlayer_ai-0.1.0-py3-none-any.whl.

File metadata

  • Download URL: humanlayer_ai-0.1.0-py3-none-any.whl
  • Upload date:
  • Size: 13.0 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.14

File hashes

Hashes for humanlayer_ai-0.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 120ff1642beda63cd625c7872c56e1dc06983a63d7a152dbb075074b0cf71a79
MD5 9acf9f65133f01d8807d454b122b613e
BLAKE2b-256 7f043f69c95de186b30506ff66213e528648eed7d18c4502dc3ccdac96ce72cc

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