Harness Engineering Python Packages

datachain

The Context Layer for unstructured data: typed, versioned datasets over S3, GCS, Azure

memtomem

Markdown-first, long-term memory infrastructure for AI agents. Hybrid BM25 + semantic search across markdown/code files via MCP.

tok-protocol

tok is an invisible bridge

interlocks

One opinionated command surface for local checks and CI, replacing per-repo quality-script sprawl with a consistent, enforceable workflow.

carl-studio

Coherence-Aware Reinforcement Learning (CARL) - breakthrough LLM post-training and test-time training paradigm. carl builds the world's most advanced and intelligent agent systems that are a step change over current gen agents

kogwistar

Graph-native AI substrate for HyperGraphRAG. Knowledge, Workflow orchestration, conversation memory and wisdom are modeled as hypergraphs. Support strong event sourced provenance and replay.

pyflue

Python-first agent harness framework with Markdown skills, sessions, sandboxes, and pluggable backends.

graphrefly

Reactive harness layer for agent workflows. Describe automations in plain language, trace every decision, enforce policies, persist checkpoints. Python. Zero dependencies.

ceo-py

An ultra-lightweight agentic AI framework based on the ReAct paradigm.

superagentic-metaharness

Meta Harness Implementation

carl-core

Coherence-Aware Reinforcement Learning (CARL) - breakthrough LLM post-training and test-time training paradigm. carl builds the world's most advanced and intelligent agent systems that are a step change over current gen agents

syrin

The Python Harness for Production AI Multi-Agent Systems

browser-automation-cli

A lightweight, self-hosted browser automation tool with a background daemon and CLI client. Tailored for agent harnesses.

pantheon-agents

A general, evolvable, and distributed agent framework & harness for data science.

repomap-cli

Codebase awareness for coding agents: tree-sitter AST maps, 13-language LSP, pre/post-edit impact analysis

agent-slides

Agent skill for generating professional PowerPoint (.pptx) presentations, powered by an agent-ready CLI and Python library

claude-md-compiler

Compile CLAUDE.md into a lockfile, then enforce repo changes against it in local runs, staged diffs, or CI.

mcpomni-connect

Open Python agent harness for production AI agents: parallel tool execution, structured observations, loop detection, memory, workspace files, MCP tools, subagents, background tasks, and OmniServe APIs.

harness0

Harness engine for AI Agents. From demo to production.

ctxtual

Context engineering library for AI agents: decorator-based workspace storage for large tool results

adam-mcp

MCP server that exposes Adam Framework vault memory as tools for Claude Desktop, Cursor, and any MCP-compatible client

autono

A ReAct-Based Highly Robust Autonomous Agent Framework

hdd-dev

HDD(Harness-Driven Development) - Keep AI-built systems coherent when requirements change.

graphrefly-py

Reactive harness layer for agent workflows. Describe automations in plain language, trace every decision, enforce policies, persist checkpoints. Python. Zero dependencies.