Harness engineering is the discipline of designing the environments, constraints, and feedback loops around AI coding agents that make them reliable at scale. The formula is: Agent = Model + Harness. Harness is everything that isn't the model: the infrastructure that governs how the agent operates, what it can access, and how it self-corrects.
In this article I'm collecting all the technologies I stumbled upon while researching harness engineering. The list is not exhaustive, but it should give you a good starting point to explore the landscape. This aims to be a living document, so I'll keep it updated.
<Link for the full technology landscape in HTML>
SDD
Spec-Driven Development
- OpenSpec - Spec-driven development for AI coding assistants
- spec-kit - Toolkit to get started with Spec-Driven Development
- BMAD - Breakthrough Method for Agile AI Driven Development
- Spec Kitty - Spec Kitty is an open-source CLI for turning product intent into a repo-native AI coding workflow:
- Specula - Specula: An agentic tool for finding deep bugs in system code using TLA+. It uses coding agents to write TLA+ specs of the target system.
- GSD Core - Context-engineering and spec-driven development framework that drives AI coding agents
- Elephant-Goldfish Model - Open-source plugin for designing and reviewing software with agents
Coding Assistants
Coding assistants usable from a terminal
General Purpose
- pi.dev - Minimal agent harness — adapt to your workflow, not vice versa
- OpenCode - OpenCode is an open source agent that helps you write code in your terminal, IDE, or desktop.
- NanoCoder - An open coding agent for your terminal, built by a community collective rather than a company. Bring your own model, keep your code on your machine, and owe nothing to anyone.
- Qwen Code - An open-source AI agent for the terminal, optimized for Qwen series models
- DeepSeek Harness - Open-source agent harness developed by DeepSeek AI
- jcode - The most RAM efficient harness
- prime-agent - A self-improving RLM agent for coding workflows and long-running autonomous tasks
- forge code - AI enabled pair programmer for Claude, GPT, O Series, Grok, Deepseek, Gemini and 300+ models.
- Claude Code - Agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows
- Codex - Lightweight coding agent that runs in your terminal
- Cursor CLI - Cursor's autonomous coding agent for the terminal, distributed via npm (cursor-agent)
- GitHub Copilot CLI - GitHub Copilot coding agent in your terminal
Code Review
- open-code-review - Hybrid-architecture code review tool combining deterministic pipelines with an LLM agent for precise line-level comments and built-in multi-language rulesets
Local-First
- Codehamr - A minimal, local-first coding agent.
- Atomic Agent - A local-first AI agent that runs on your machine, with local or cloud models.
SDLC Templates or Template Management
Holding agentic configurations, skills, etc.
- OpenViking - Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.
- Mimir - Mimir is a foundation for self-evolving methodologies for well-defined tasks
- ArcKit - ArcKit is an enterprise Architecture governance harness that wraps your AI coding assistant
- Hivemind - Self-hosted multi-agent AI team platform. 40+ built-in tools, 150+ agent templates.
- OpenHands - The self-hosted developer control center for coding agents and automations.
- Fabric - Open-source framework for augmenting humans with AI; a crowdsourced, modular set of AI prompt skills usable anywhere
Orchestration
General Purpose
- Symphony - Isolated autonomous implementation runs for teams
- Conductor - Multi-agent workflows with GitHub Copilot SDK
- OpenSquilla - Token-efficient AI agent with higher intelligence density
- Fabro - Version-controlled workflow graphs orchestrating AI agents, commands, and human gates
- smolagents - Barebones library for agents that think in code
- OpenHive - Multi-agent harness for production AI workloads
- bernstein - Audit-grade multi-agent orchestration, HMAC-chained audit log
- Scion - Scion is a container-based orchestration platform for running multiple LLM deep agents concurren
- AWS AI-DLC - AI-Driven Life Cycle (AI-DLC) adaptive workflow steering rules for AI coding agents
- HumanLayer - HumanLayer runs coding agents (Claude Code, Codex, OpenCode, Pi) on your machines and lets your team steer them from one place: tasks, sessions, plans, and reviews, synced between a desktop app, a web app, and a daemon on each host.
- Harness² - Recursive agent harnessing framework that optimizes the system around a model: prompts, tools, memory, skills and orchestration
Multi-Agent Workspaces
- T3 Code - T3 Code is an agent harness control surface.
- Paperclip - The open-source app everyone uses to manage agents at work
- Orca - Run a fleet of coding agents (Claude Code, Codex) with your own subscription; available on desktop, mobile and remote runtime
AST and Code Parsers
- CocoIndex Code - AST-based lightweight code search engine CLI; saves 70% tokens
- CodeGraphContext - MCP server indexing local code into a graph database
- ast-grep - CLI tool for code structural search, lint and rewriting in Rust
- ripwire - The ripgrep of AI context: a zero-dependency C++23 CLI + MCP server for coding agents.
- Graphify - Turn code and docs into queryable AI knowledge graphs
- Understand Anything - Turn any codebase, knowledge base, or docs into an interactive knowledge graph you can explore, search, and ask questions about.
- Trail-Code (Graft) - Self-learning context layer for coding agents; tree-sitter code map plus shared team learnings delivered via MCP
- Serena - MCP toolkit for coding with semantic retrieval and editing; the IDE for your agent
- enola - One graph of your whole software system, built from the code on your machine; parses modules, functions, API routes and DB tables across repos, exposed to agents over MCP with CI checks for architectural regressions
Sandboxes
- NVIDIA OpenShell - Sandboxed agent runtime with hardware-enforced isolation and policy
- forkd - fork() for AI agent microVMs; spawn 100 children in ~100ms
- OpenSandbox - Secure, fast, extensible sandbox runtime for AI agents
- Daytona - Elastic infrastructure for running AI-generated code; sub-90ms creation
- E2B - E2B provides isolated sandboxes that let agents safely execute code, process data, and run tools.
- Herdr - Native app for running coding agents; your agent reshapes the UI, and herdr-core keeps sessions alive when the app is closed
- Cove - Hosted agent infrastructure by GitGuardian; every run boots an isolated sandbox with your own disk, scales to thousands of copies and handles OAuth for Slack, GitHub, Stripe and MCP
- Microsandbox - Easy, fast, programmable and local-first microVM runtime and library
- StrandsAgents box - Rust sandbox for AI agents; OS isolation with default-deny Dogwood policies and credential injection that keeps secrets outside the agent, on macOS and Linux
Working with Skills
Skill Reference
- Agent Skills - Standardized open format for extending AI agent capabilities via SKILL.md files
Skill Syntax Validation
- skill-validator - Validates skill content against the Agent Skill specification
Skill Repositories
- OSMT - OSMT is the Open Skills Management Tool
- SkillHub - Self-hosted, open-source agent skill registry for enterprises.
Skill Dependency Management
- skills (Vercel Labs) - Open agent skills CLI; supports OpenCode, Claude Code, Codex, Cursor and 68+ more
- Microsoft APM (Agent Package Manager - An open-source, community-driven dependency manager for AI agents.
- skills-manager - A lightweight desktop app to manage, sync, and organize AI agent skills across 50+ coding tools — Claude Code, Codex, Cursor, Copilot, Gemini CLI, and more.
- Microsoft waza - CLI / framework for Agent Skills — create, test, measure and improve skill quality and effectiveness.
Skill Security Scanners
- DefenseClaw - Security governance for agentic AI; scan capabilities, inspect traffic, audit evidence
- SkillSpector - Security scanner for AI agent skills; detects vulnerabilities before installation
- Cloudflare - security audit skill - Coding-agent skill for multi-phase security audits with independently verified, machine-readable findings
Knowledge Base
LLM Wikis
- OpenKnowledge - Beautiful, AI-native markdown IDE and LLM wiki
- pi-llm-wiki - Self-maintaining, Obsidian-compatible knowledge base for pi
- Khoj - Self-hostable AI second brain; answers from the web or your docs, custom agents and scheduled automations
Open Knowledge Format (OKF)
- OKF Ecosystem - Open Knowledge Format ecosystem; tools, spec and docs for AI agent knowledge bases
- okflint - Deterministic compliance linter for OKF bundles; profile-based rule enforcement
- okf CLI - CLI for git-based OKF knowledge bases; keeps bundles fresh via git hooks
- OpenWiki - OpenWiki is a CLI that writes and maintains agent documentation for your codebase in OKF compatible format.
Technology Specifics
- Context7 - Up-to-date documentation for LLMs and AI code editors
- LangChain docs MCP + skills - Use LangChain Docs Programmatically with MCP and installed skills
- NVIDIA Agent Skills - Official NVIDIA-verified skills catalog for CUDA-X, AI Blueprints and platform tools
- DataHub Skills - Agent skills for working with DataHub
- Angular skills - Official Angular agent skills package; teaches AI coding assistants how to build Angular apps.
Token Control
- rtk - CLI proxy reduces LLM token consumption by 60-90% on common dev commands; single Rust binary
- ponytail - Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
- Headroom Desktop - MacOS menu bar app cuts Claude Code and Codex token costs by ~50%
- Caveman - An efficiency stack for AI agents - cuts 65% of tokens by talking like caveman
Observability
- Langfuse - Open source AI observability platform; LLM evals, metrics, tracing, prompt management
- Phoenix - AI observability and evaluation by Arize; traces, evals, datasets
- Claude Code - OTEL - How to enable and configure OpenTelemetry for Claude Code.
- SigNoz - Coding agent observability - ClaudeCode observability with SigNoz and OpenTelemetry
- agentops - Python SDK for AI agent monitoring, LLM cost tracking, benchmarking
- Tokscale - High-performance CLI and dashboard for tracking token usage and costs across multiple AI coding agents, with a global leaderboard
References
Thoughtworks
- https://martinfowler.com/articles/harness-engineering.html
- https://martinfowler.com/articles/exploring-gen-ai.html
- https://martinfowler.com/bliki/FutureOfSoftwareDevelopment.html
Whitepapers from Google
- https://www.kaggle.com/whitepaper-the-new-SDLC-with-vibe-coding
- https://www.kaggle.com/whitepaper-agent-skills
- https://www.kaggle.com/whitepaper-vibe-coding-agent-security-and-evaluation
- https://www.kaggle.com/whitepaper-spec-driven-production-grade-development-in-the-age-of-vibe-coding
- https://www.kaggle.com/whitepaper-agent-tools-and-interoperability
- https://www.kaggle.com/whitepaper-context-engineering-sessions-and-memory

Top comments (0)