{"data":{"slug":"walkinglabs-learn-harness-engineering","name":"learn-harness-engineering","tagline":"Harness engineering beginner tutorial from ground up","github_url":"https://github.com/walkinglabs/learn-harness-engineering","owner":"walkinglabs","repo":"learn-harness-engineering","owner_avatar_url":"https://avatars.githubusercontent.com/u/271878417?v=4","primary_language":"TypeScript","stars":12330,"forks":1295,"topics":["agent","agentic","agentic-ai","ai","ai-agent","dsh","dsh-plugin","harness","harness-engineering","harness-framework","llm"],"archived":false,"github_pushed_at":"2026-08-19T09:42:52+00:00","maintenance_label":"Very active","stars_delta_30d":1815,"url":"https://www.graphcanon.com/tools/walkinglabs-learn-harness-engineering","markdown_url":"https://www.graphcanon.com/tools/walkinglabs-learn-harness-engineering.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/walkinglabs-learn-harness-engineering","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=walkinglabs-learn-harness-engineering","description":"Harness engineering beginner tutorial, from 0 to 1","homepage_url":"https://walkinglabs.github.io/learn-harness-engineering/","license":"MIT","open_issues":6,"watchers":57,"ai_summary":"Provides structured guidance for leveraging AI agents in software development through harnessing with TypeScript.","readme_excerpt":"## Quick Start: Improve Your Agent Today\n\nYou don't need to read all 14 lectures before you start getting value. If you're already using a coding agent on a real project, here's how to improve it right now.\n\nThe idea is simple: instead of just writing prompts, give your agent a set of structured files that define what to do, what's been done, and how to verify the work. These files live inside your repo, so every session starts from the same state.\n\n```text\n    YOUR PROJECT ROOT\n    ├── AGENTS.md              <-- the agent's operating manual\n    ├── CLAUDE.md              <-- (alternative, if using Claude Code)\n    ├── init.sh                <-- runs install + verify + start\n    ├── feature_list.json      <-- what features exist, which are done\n    ├── claude-progress.md     <-- session progress (historical filename; agent-agnostic)\n    └── src/                   <-- your actual code\n```\n\nGrab the starter templates from the [Resource Library](https://walkinglabs.github.io/learn-harness-engineering/en/resources/) and drop them into your project. That's it. Four files, and your agent sessions will already be significantly more stable than running on prompts alone.\n\n`claude-progress.md` is a generic, repository-local session progress log; the\nname is retained for compatibility with the course examples. It is not tied to\nClaude Code and is not updated automatically by any agent. Codex, OpenHands,\nAntigravity, and other coding agents can use the same file when their root\ninstructions tell them to read it at startup and update it before handoff.\n\n---\n\n---\n\n## Requirements\n\nThis is a course where you actually run coding agents.\n\nYou need at least one of these tools:\n\n- Claude Code\n- Codex\n- Another IDE or CLI coding agent that supports file editing, command execution, and multi-step tasks\n\nThe course assumes you can:\n\n- Open a local repository\n- Allow the agent to edit files\n- Allow the agent to run commands\n- Inspect output and re-run tasks\n\nIf you don't have such a tool, you can still read the course content, but you won't be able to complete the projects as intended.\n\n---","github_created_at":"2026-03-29T12:28:19+00:00","created_at":"2026-07-07T17:39:30.125217+00:00","updated_at":"2026-08-19T12:01:11.815332+00:00","categories":[{"slug":"ai-agents","name":"AI Agents","url":"https://www.graphcanon.com/categories/ai-agents","markdown_url":"https://www.graphcanon.com/categories/ai-agents.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/ai-agents"},{"slug":"developer-tools","name":"Developer Tools","url":"https://www.graphcanon.com/categories/developer-tools","markdown_url":"https://www.graphcanon.com/categories/developer-tools.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/developer-tools"}],"tags":[{"slug":"agent","name":"agent"},{"slug":"agentic-ai","name":"agentic-ai"},{"slug":"coding-agent","name":"coding-agent"},{"slug":"tutorial","name":"tutorial"},{"slug":"typescript","name":"typescript"}],"trust":{"provenance":{"is_fork":false,"github_id":1195243030,"owner_type":"Organization","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-08-19T12:01:10.931Z","maintenance":{"label":"Very active","score":96,"methodology":"github_public_v1","releases_90d":0,"days_since_push":0,"last_release_at":null,"stars_delta_30d":1815,"open_issues_delta_30d":-8},"security_summary":{"status":"ok","scanner":"osv@v1","low_count":0,"high_count":0,"last_scan_at":"2026-07-11T11:14:39.003Z","medium_count":0,"scan_profile":"deps","critical_count":0}},"capability_facts":{"mcp":{"source":"repo_scan","observed_at":"2026-08-19T12:01:11.396Z","server_manifest":false},"scan":{"source":"repo_scan","observed_at":"2026-08-19T12:01:11.396Z"},"languages":{"value":["typescript","javascript"],"source":"github.language+package.json","observed_at":"2026-08-19T12:01:11.396Z"},"license_spdx":{"value":"MIT","source":"github.license","observed_at":"2026-08-19T12:01:11.396Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":null,"constraints":null,"when_to_use":["- **You are new to utilizing coding agents and want structured guidance:** The tool starts from the basics, making it ideal if you're just beginning to use or plan to integrate coding agents.","- **When your project could benefit from increased automation guided by an AI agent:** By structuring files that dictate what tasks should be done, what has been completed, and how work can be checked","- **Your team is looking for stability in handling AI-driven sessions:** Learn-Harness-Engineering offers immediate benefits to stabilize the use of coding agents, making it a solid choice if your aim","- **You have access to tools like Claude Code or Codex:** The tool requires specific capabilities and these platforms offer the necessary functionalities to follow the course and maximize the learning"],"when_not_to_use":["- **When you already possess advanced knowledge in coding agent integration:** This tutorial is geared towards beginners, and an experienced user might find the level of detail excessive or basic.","- **If your project doesn't require structured guidance for AI-driven tasks:** Learn-Harness-Engineering focuses heavily on creating files that guide agents which may be unnecessary if your setup can","- **You lack access to coding agent tools with multi-step task management capabilities:** The course assumes the availability of these specific types of tools, so without them you won't fully benefit"],"source":"enrich:decision_facts","observed_at":"2026-07-11T15:55:28.386Z"},"constraint_facets":null,"decision_summary":[{"label":"Adopt for","value":"Learn-Harness-Engineering is a TypeScript-based tutorial designed for beginners in harness engineering focused on integrating AI agents into software development workflows efficiently."}]}}