MoonshotAI
Featured
Kimi-K3
Kimi K3 is an open-weight, 2.8T-parameter native multimodal agentic model with a 1M-token context window, designed for frontier coding, knowledge work, and reasoning tasks.
harness-engineering
Harness Engineering is a methodology for improving coding agent outputs by carefully crafting the environment around them—providing curated context, tools, and executable constraints that encode an organization’s nonfunctional requirements and cumulative lessons.
Stars
2,390
7-day growth
+176
Forks
245
Open issues
2
License
CC-BY-4.0
Last updated
2026-07-18
AI repository intelligence
FR-AI / ANALYSISWhy it is worth attention
It introduces a systematic, evidence-based framework for making organizational knowledge and quality standards recoverable by agents, turning agent interaction into a cumulative feedback loop rather than relying on model weights alone.
Who it is for
- Software engineers using AI coding agents
- Engineering managers overseeing agent-integrated workflows
- Platform or infrastructure engineers building agent tooling
- Organizations adopting agent-based development processes
Use cases
- Setting up a repository so that a coding agent automatically understands project conventions and nonfunctional requirements
- Embedding reliability, security, and maintainability checks as executable constraints in an agent’s environment
- Building a feedback loop from accepted work and failures to continuously improve agent behavior
- Implementing ‘last-mile deployment’ that supplies context, capabilities, and authority without requiring human review of every agent action
Strengths
- Clearly defined concept with concrete levers (context and tools) and a systems-level framing
- Includes practical playbooks, a thesis index, and a reference implementation (AGENTS.md)
- Draws on real-world sources and prior work, showing external validation and community engagement
- Focuses on cumulative coherence and organizational learning, making the practice scalable
Considerations
- Requires significant upfront investment to curate the harness for a specific organization or system
- Assumes a capable black-box agent; the framework may not address limitations of weaker models
- Organization-specific process data must be manually captured and maintained, posing a maintenance burden
README quick start
Harness Engineering
“Most people do not know that they can just point their agents at my writing, tweets, podcasts, and talks and improve the output of their agents by 100x.”
Harness engineering, the practice of improving agent output by shaping the environment around it, holds a chosen model and coding agent constant as a black box. It improves the two external levers—context and tools—and curates the environment around them. The worker should be able to recover intent, operate the real system, respect authority, prove the outcome, and leave the next run better equipped.
A central purpose of that environment is to carry an organization's nonfunctional requirements: the quality attributes and constraints governing reliability, security, compatibility, maintainability, performance, operability, risk posture, and polish. The harness also carries local decisions about how to prioritize, trade off, and satisfy those requirements. Ryan adopted a systems-level framing from 2026’s [un]prompted conference that describes this as getting the whole universe of nonfunctional requirements into code. Make the Repository Teach the Agent develops how the requirements and decisions become retrievable context, examples, tools, and executable constraints.
Because work is an iterative game, a harness can make organizational judgment cumulative. Lessons from accepted work, corrections, failures, and user responses become context, boundaries, tools, examples, and checks that shape later trajectories. Over time, that feedback loop can make coherence cumulative across agent-maintained artifacts.
[Code is how an agent uses a computer]. That internal action language can produce reliable domain outcomes for people who never review the implementation when [last-mile deployment] supplies the organization’s context, capabilities, authority, and proof.
[Code is how an
Description
🐎 Ryan Lopopolo’s anthology, field guide, and agent context bundle for harness engineering
Related repositories
Similar projects matched by category, topics, and programming language.
mshumer
Featured
Claude-of-Duty
A browser-based first-person shooter built entirely with procedural generation and orchestrated AI agents, featuring 55k lines of Three.js/WebGL2 code and no art assets.
Jakubantalik
Featured
thinking-orbs
A React component library that renders six hand-tuned animated thought orb loading indicators on a plain 2D canvas, with two purpose-tuned sizes and automatic theme detection for AI and agent UIs.