Skip to main content

Pre-configured Haive agent presets and templates for common use cases

Project description

haive-prebuilt

PyPI version Python Versions License: MIT CI Docs PyPI Downloads

Pre-configured Haive agent presets and templates for common use cases.

40+ ready-to-deploy agents for specific business domains: customer support, research, content generation, code review, and more. Built on the Haive framework with sensible defaults — drop in and use, or customize as needed.


Why haive-prebuilt?

The Haive framework gives you the building blocks. haive-prebuilt gives you the finished products. Each agent is:

  • Pre-tuned — Temperature, system message, and tools selected for the use case
  • Domain-specific — Configured for a particular industry or task
  • Production-ready — Tested with real LLMs (no mocks)
  • Customizable — Override any setting via the standard Haive config pattern

Use haive-prebuilt when you need an agent for a common use case and don't want to start from scratch. Use the foundation packages (haive-core, haive-agents) when you need something custom.


Available Agents

Customer Support

  • Customer Support Agent — Multi-turn conversation with escalation
  • FAQ Bot — Document-grounded Q&A
  • Ticket Triage Agent — Classify and route support tickets

Research & Analysis

  • Market Research Agent — Web search + synthesis for market analysis
  • Competitive Analysis Agent — Compare products/companies
  • Trend Analyst — Identify and report on trends

Content Generation

  • Blog Writer — Long-form content with research
  • Social Media Manager — Posts for Twitter, LinkedIn, etc.
  • Email Drafter — Professional email drafting
  • Newsletter Generator — Curated weekly newsletters

Code & Development

  • Code Reviewer — Reviews PRs and suggests improvements
  • Bug Triage Agent — Classify and prioritize bug reports
  • Documentation Generator — Auto-generate docstrings and READMEs

Data & Analytics

  • Data Analyst — Pandas + SQL for data exploration
  • Report Generator — Convert data into narrative reports
  • Anomaly Detector — Identify outliers in datasets

Domain-Specific

  • Legal Document Reviewer
  • Medical Note Summarizer
  • Financial Analyst
  • HR Resume Screener

Quick Start

from haive.prebuilt import get_agent

# Get a pre-configured agent
agent = get_agent("customer_support")
result = agent.run("How do I reset my password?")

# Or instantiate directly
from haive.prebuilt.customer_support import CustomerSupportAgent
agent = CustomerSupportAgent(
    company_name="Acme Corp",
    knowledge_base_url="https://docs.acme.com",
)

Customization

Every prebuilt agent uses standard Haive patterns. Override any config:

from haive.prebuilt.research import MarketResearchAgent
from haive.core.engine.aug_llm import AugLLMConfig

# Override the engine
agent = MarketResearchAgent(
    engine=AugLLMConfig(
        model="claude-opus-4-6",
        temperature=0.5,
        system_message="Custom instructions...",
    ),
    max_iterations=10,
)

Installation

pip install haive-prebuilt

Documentation

📖 Full documentation: https://pr1m8.github.io/haive-prebuilt/


Related Packages

Package Description
haive-core Foundation: engines, graphs
haive-agents Production agent implementations
haive-tools Tool implementations

License

MIT © pr1m8

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

haive_prebuilt-1.0.1.tar.gz (189.2 kB view details)

Uploaded Source

Built Distribution

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

haive_prebuilt-1.0.1-py3-none-any.whl (284.7 kB view details)

Uploaded Python 3

File details

Details for the file haive_prebuilt-1.0.1.tar.gz.

File metadata

  • Download URL: haive_prebuilt-1.0.1.tar.gz
  • Upload date:
  • Size: 189.2 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.12.11

File hashes

Hashes for haive_prebuilt-1.0.1.tar.gz
Algorithm Hash digest
SHA256 1bda6c62595599065791a5c09b70d50961d71fdf52b13ef14f94257b12df896e
MD5 8273e387d817a362765f3906b0787833
BLAKE2b-256 5735f3e0f6c19c14b265988ee4e74cbe5a7d3871bdc22fbee031af3231e0a782

See more details on using hashes here.

File details

Details for the file haive_prebuilt-1.0.1-py3-none-any.whl.

File metadata

  • Download URL: haive_prebuilt-1.0.1-py3-none-any.whl
  • Upload date:
  • Size: 284.7 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.12.11

File hashes

Hashes for haive_prebuilt-1.0.1-py3-none-any.whl
Algorithm Hash digest
SHA256 5bd9c1a6fb8d55c18224eaf21e50070ac710008625145525299ac4a32e304df5
MD5 410bd3913204caf0ff795811d50d75a1
BLAKE2b-256 3ad668295511eb3dabc5a04c343c60029fc3aeda114dc77553b781fd52b62eb7

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