Skip to content
Back to search

better-harness

MarketplaceWorks with Claude CodeGitHub

Better Harness plugins for AI delivery readiness.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today

Check it deeper on your machine

It has changed since its scan on 11 Nov 2026. Its scripts were only partly scanned.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/QoderAI/better-harness with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/QoderAI/better-harness'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/qoderai-better-harness/badge.svg)](/listings/qoderai-better-harness)

HTML

<a href="/listings/qoderai-better-harness"><img src="/api/listings/qoderai-better-harness/badge.svg" alt="findsafeskills safety rating" /></a>

Install

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Installation differs by coding agent. Install Better Harness separately for each host, except that Qoder CLI can use the version bundled with Qoder Desktop. After installing or updating a plugin, start a new session or task so the host reloads its plugin inventory.

Claude Code

Register this repository as a Claude Code marketplace:

/plugin marketplace add QoderAI/better-harness

Then install Better Harness:

/plugin install better-harness@better-harness

Verify discovery from the shell:

claude plugin details better-harness@better-harness

The details should include `Skills (1) better-harness`. Then start a new Claude session in the repository you want to analyze and run the report prompt:

/better-harness analyze this project's AI coding workflow and generate an evidence-backed report

Claude Code defaults to a self-contained `report.html` with paired `report.md` and `findings.json` under the repository's `.claude/better-harness` report root. Ask for inline or no-files output to keep the result in chat only. Workspace- matching local Claude sessions are included when available; missing evidence stays explicit rather than being inferred.

Codex

<a id="codex-desktop"></a>

Codex Desktop

1. Open **Settings > Plugins**. 2. Select **+ Add > From Marketplace**. 3. Enter the Git repository URL, set its Git ref, and leave **Sparse paths** empty for this single-plugin repository. 4. Select **Add marketplace**, then install **Better Harness** from the new marketplace. 5. Start a new task in the repository you want to analyze and run the report prompt:

@better-harness analyze this project's AI coding workflow and generate an evidence-backed report

Use `https://github.com/QoderAI/better-harness.git` with Git ref `main`.

<a id="codex-cli"></a>

Codex CLI

Add the repository source:

codex plugin marketplace add \
  'https://github.com/QoderAI/better-harness.git' \
  --ref main

Then inspect and install Better Harness:

codex plugin list --marketplace better-harness
codex plugin add better-harness@better-harness

Start a new Codex task in the repository you want to analyze and run the report prompt:

$better-harness:better-harness analyze this project's AI coding workflow and generate an evidence-backed report

Use the repository URL with `marketplace add`, not a raw `marketplace.json` URL. Current Codex builds use `plugin add` and `--marketplace`; examples that use `plugin install` or `--source` target a different CLI contract.

Qoder

Better Harness is built into the [Qoder](https://qoder.com/) desktop app, so no Marketplace or local plugin installation is required there. Choose either entry point:

1. **From a session:** Open the repository you want to analyze, start a new session, and run the report prompt:

   /better-harness analyze this project's AI coding workflow and generate an evidence-backed report

2. **From Quest (Qoder 1.18.0+):** Open Quest, then select **Better Harness (Beta)** from the left sidebar.

Qoder CLI

If Qoder Desktop is installed, Better Harness is already available in Qoder CLI. No marketplace or plugin installation is required. Start a new Qoder CLI session in the repository you want to analyze and run the report prompt:

/better-harness analyze this project's AI coding workflow and generate an evidence-backed report

Only when using Qoder CLI without Qoder Desktop, inspect the current manual installation disposition before following:

From marketplace

# Add the plugin marketplace source
qodercli plugin marketplace add 'https://github.com/QoderAI/better-harness.git'

# Install the plugin
qodercli plugin install better-harness@better-harness

# Check installation
qodercli plugin list

From git

# Make sure directory exist
mkdir -p $HOME/.qoder/plugins/marketplaces/

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 20 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. {
  2. "name": "better-harness",
  3. "description": "Better Harness plugins for AI delivery readiness.",
  4. "owner": {
  5. "name": "Qoder",
  6. "email": "dev@qoder.com"
  7. },
  8. "plugins": [
  9. {
  10. "name": "better-harness",
  11. "description": "Build an AI-ready engineering system for safe coding-agent delivery and continuous software improvement.",
  12. "version": "0.7.0-alpha2",
  13. "source": "./",
  14. "author": {
  15. "name": "Qoder",
  16. "email": "dev@qoder.com"
  17. }
  18. }
  19. ]
  20. }
README.md · README · 481 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. <p align="center">
  2. <img src="assets/logo.svg" alt="Better Harness logo" width="56" height="56">
  3. </p>
  4. <h1 align="center">Better Harness</h1>
  5. <p align="center">
  6. English · <a href="README.zh-CN.md">简体中文</a>
  7. </p>
  8. <p align="center">
  9. <strong>Delegate coding to agents. Improve the loop around them.</strong>
  10. </p>
  11. <p align="center">
  12. Better Harness provides open-source insights for the Agent Work Loop. It runs
  13. through your Coding Agent and turns project and session evidence into prioritized
  14. improvements and verifiable next steps. Missing evidence stays explicit.
  15. </p>
  16. <p align="center">
  17. <a href="https://www.npmjs.com/package/@qoder-ai/better-harness"><img src="https://img.shields.io/npm/v/@qoder-ai/better-harness.svg" alt="npm version"></a>
  18. <a href="LICENSE"><img src="https://img.shields.io/badge/license-MIT-blue.svg" alt="MIT License"></a>
  19. </p>
  20. <p align="center">
  21. <a href="https://qoderai.github.io/better-harness/?utm_source=github&utm_medium=referral&utm_campaign=repository_landing&utm_content=readme_hero">Website</a> ·
  22. <a href="#quick-start">Choose your host</a> ·
  23. <a href="#see-it-in-action">Sample report</a> ·
  24. <a href="https://qoderai.github.io/better-harness/docs/introduction">Docs</a>
  25. </p>
  26. ## Quick start
  27. Analyze and improve your coding workflow with: [Claude Code](#claude-code), [Codex Desktop](#codex-desktop), [Codex CLI](#codex-cli), [Qoder Desktop/CLI](#qoder), [Cursor](#cursor), or [GitHub Copilot CLI](#github-copilot).
  28. Choose the host you already use to get its exact installation, verification,
  29. invocation, and report-output steps. Better Harness does not use one universal
  30. entrypoint across every host.
  31. This README shows inline setup for the most common hosts. Additional supported
  32. hosts (Qwen Code, Pi, Kimi Code, WorkBuddy, and Grok) keep their steps and
  33. boundaries in the [installation guide](docs/docs/installation.mdx) and the
  34. [public Host Adapter Matrix](docs/docs/hosts/adapter-matrix.md); see
  35. [More adapters](#more-adapters). README placement is a display choice, not a
  36. support-level claim.
  37. Better Harness scopes behavior claims to relevant Task Episodes and the
  38. surrounding project mechanisms. Qoder and Cursor produce host-native Canvas
  39. reports; Claude Code, Codex, Qwen Code, GitHub Copilot, and Kimi Code produce
  40. self-contained HTML with paired Markdown. Missing or partial evidence remains
  41. explicit. See the [Host Adapter Matrix](docs/adapters/README.md) for current
  42. coverage and output differences.
  43. ## See it in action
  44. The report keeps missing evidence explicit and turns supported gaps into
  45. prioritized findings with an impact, expected output, scoped repair, and
  46. acceptance checks.
  47. <p align="center">
  48. <a href="https://qoderai.github.io/better-harness/demo/better-harness-report/"><img src="assets/demo/better-harness-findings-report.png" alt="Better Harness HTML report showing an evidence-bounded finding with its impact, expected output, scoped AI fix, and acceptance checks" width="900"></a>
  49. </p>
  50. <p align="center">
  51. <sub><a href="https://qoderai.github.io/better-harness/demo/better-harness-report/">Open the complete self-contained English HTML report</a>
  52. (<a href="assets/demo/better-harness-report.html">source</a>).</sub>
  53. </p>
  54. For delivery tracing, the interactive [Harness Inspector](https://qoderai.github.io/better-harness/inspector)
  55. follows product intent through agent activity, sessions, files, and commits in
  56. a read-only workspace, keeping evidence strength and limitations visible:
  57. <p align="center">
  58. <a href="https://qoderai.github.io/better-harness/inspector"><img src="docs/assets/harness-inspector/session-view.png" alt="Harness Inspector session view: a synchronized timeline of prompts, tool calls, and commits with the Evidence Drawer explaining each link" width="900"></a>
  59. </p>
  60. <p align="center">
  61. <sub><a href="https://qoderai.github.io/better-harness/inspector">Open the interactive Harness Inspector sample</a> (fictional English data; it never reads your workspace).</sub>
  62. </p>
  63. After you have comparable reports over time, the history view shows how the five
  64. Agent Work Loop dimensions move:
  65. <p align="center">
  66. <a href="dev/terminal-demo/README.md"><img src="assets/demo/twenty-history.png" alt="Static final frame of Better Harness report history showing five Agent Work Loop dimensions over time" width="900"></a>
  67. </p>
  68. The static final frame summarizes historical Harness reports. It shows recorded
  69. trends, not causal proof of improvement. [See how the demo was recorded](dev/terminal-demo/README.md).
  70. ## Why Better Harness?
  71. AI coding agents change code fast, but the workflow around them is often the
  72. weak point:
  73. - 🎯 **Fuzzy goals** — the agent confidently solves the wrong problem.
  74. - 🧭 **Improvised steps** — work happens on paths nobody can reproduce.
  75. - ✅ **"It works" without proof** — validation is incomplete or missing.
  76. - 🚢 **Speed over safeguards** — review and delivery checks get bypassed.
  77. - 🧠 **Lessons lost** — the same friction comes back on the next task.
  78. Reviewing only the final diff misses these system-level problems. Better Harness
  79. analyzes the workflow around the diff: it gathers project evidence (and session
  80. evidence where supported), evaluates five connected dimensions, and turns
  81. concrete gaps into prioritized findings — each tied to its evidence, expected
  82. outcome, repair boundary, and validation route, so a team can improve one issue
  83. at a time.
  84. ## How Better Harness works
  85. Better Harness uses a
  86. [feedforward-and-feedback](https://martinfowler.com/articles/harness-engineering.html#FeedforwardandFeedback)
  87. loop that combines guidance available before work starts with signals available
  88. after the agent acts:
  89. - **Feedforward guides** — `AGENTS.md`, specs, Skills, and acceptance criteria
  90. steer the agent before it acts.
  91. - **Feedback sensors** — linters, tests, Hooks, and evaluation agents observe results
  92. and help the agent self-correct.
  93. Across that loop, it evaluates five parts of delivery — the **Agent Work Loop**:
  94. [![Agent Work Loop: five dimensions from task understanding through learning capture](assets/agent-work-loop-en.svg)](models/agent-work-loop.md)
  95. | Dimension | The question it answers | Backed by |
  96. | --- | --- | --- |
  97. | **Task Understanding** | Does the agent know the goal and what "done" means? | Rules, `AGENTS.md`, specs, `DESIGN.md` |
  98. | **Controlled Execution** | Is the work on supported, repeatable paths? | Skills, commands, MCP tools, sandbox boundaries |
  99. | **Change Validation** | Is there evidence the change actually works? | Tests, lint, Hooks, observable diagnostics |
  100. | **Reliable Delivery** | Does AI speed bypass quality checks or acceptance? | Human review, approvals, CI/CD, recovery paths |
  101. | **Learning Capture** | Does the next task benefit from this one? | Loop Discovery, reusable SDLC Skills, Memory |
  102. Running `/better-harness` establishes a task-bounded baseline and, depending on
  103. the host, produces a visual report, a Markdown report, or both. The report
  104. combines the five-part overview, prioritized findings, detected agent assets,
  105. and an evidence brief. Each finding includes a repair action that drafts a
  106. scoped fix plan for review.
  107. Better Harness is deliberately honest: unobserved behavior stays explicit instead
  108. of becoming an unsupported score or claim. Passing a current check proves that
  109. the intervention was exercised; only a comparable later result can prove that
  110. the loop improved.
  111. ## What is open
  112. Better Harness opens three connected layers, not only a slash-command prompt:
  113. - **Engineering practices** — evidence and judgment guidance across
  114. [Session Evidence, Project Harness, Agent Customize, and Loop Engineering](references/README.md).
  115. - **Evaluation model** — the task-centered
  116. [Agent Work Loop](models/agent-work-loop.md), including evidence states,
  117. findings, scoring boundaries, and longitudinal validation.
  118. - **Runnable implementation** — the canonical
  119. [`/better-harness` workflow](skills/better-harness/SKILL.md), evidence
  120. collectors, analyzers, renderers, and thin
  121. [host adapters](docs/adapters/README.md).
  122. The three layers share the same boundary: configured assets can establish that
  123. a mechanism exists, but only linked task evidence can establish that it was used
  124. or improved an outcome.
  125. ## Architecture
  126. [![Better Harness architecture: host integration, three independent evidence agents, unified analysis by one lead agent, findings, host outputs, and repair](assets/better-harness-architecture-en.svg)](docs/ARCHITECTURE.md)
  127. The architecture keeps the three evidence domains independent until unified
  128. analysis by the lead agent. Every result retains a visible evidence source,
  129. owner, and validation route.
  130. ## Installation
  131. Installation differs by coding agent. Install Better Harness separately for
  132. each host, except that Qoder CLI can use the version bundled with Qoder Desktop.
  133. After installing or updating a plugin, start a new session or task so the host
  134. reloads its plugin inventory.
  135. ### Claude Code
  136. Register this repository as a Claude Code marketplace:
  137. ```text
  138. /plugin marketplace add QoderAI/better-harness
  139. ```
  140. Then install Better Harness:
  141. ```text
  142. /plugin install better-harness@better-harness
  143. ```
  144. Verify discovery from the shell:
  145. ```bash
  146. claude plugin details better-harness@better-harness
  147. ```
  148. The details should include `Skills (1) better-harness`. Then start a new Claude
  149. session in the repository you want to analyze and run the report prompt:
  150. ```text
  151. /better-harness analyze this project's AI coding workflow and generate an evidence-backed report
  152. ```
  153. Claude Code defaults to a self-contained `report.html` with paired `report.md`
  154. and `findings.json` under the repository's `.claude/better-harness` report root.
  155. Ask for inline or no-files output to keep the result in chat only. Workspace-
  156. matching local Claude sessions are included when available; missing evidence
  157. stays explicit rather than being inferred.
  158. ### Codex
  159. <a id="codex-desktop"></a>
  160. #### Codex Desktop
  161. 1. Open **Settings > Plugins**.
  162. 2. Select **+ Add > From Marketplace**.
  163. 3. Enter the Git repository URL, set its Git ref, and leave **Sparse paths**
  164. empty for this single-plugin repository.
  165. 4. Select **Add marketplace**, then install **Better Harness** from the new
  166. marketplace.
  167. 5. Start a new task in the repository you want to analyze and run the report
  168. prompt:
  169. ```text
  170. @better-harness analyze this project's AI coding workflow and generate an evidence-backed report
  171. ```
  172. Use `https://github.com/QoderAI/better-harness.git` with Git ref `main`.
  173. ![Codex Add plugin marketplace dialog with repository, Git ref, and optional sparse paths](assets/install/codex-add-marketplace.jpg)
  174. <a id="codex-cli"></a>
  175. #### Codex CLI
  176. Add the repository source:
  177. ```bash
  178. codex plugin marketplace add \
  179. 'https://github.com/QoderAI/better-harness.git' \
  180. --ref main
  181. ```
  182. Then inspect and install Better Harness:
  183. ```bash
  184. codex plugin list --marketplace better-harness
  185. codex plugin add better-harness@better-harness
  186. ```
  187. Start a new Codex task in the repository you want to analyze and run the report
  188. prompt:
  189. ```text
  190. $better-harness:better-harness analyze this project's AI coding workflow and generate an evidence-backed report
  191. ```
  192. Use the repository URL with `marketplace add`, not a raw `marketplace.json`
  193. URL. Current Codex builds use `plugin add` and `--marketplace`; examples that
  194. use `plugin install` or `--source` target a different CLI contract.
  195. ### Qoder
  196. Better Harness is built into the [Qoder](https://qoder.com/) desktop app, so no
  197. Marketplace or local plugin installation is required there. Choose either
  198. entry point:
  199. 1. **From a session:** Open the repository you want to analyze, start a new
  200. session, and run the report prompt:
  201. ```text
  202. /better-harness analyze this project's AI coding workflow and generate an evidence-backed report
  203. ```
  204. 2. **From Quest (Qoder 1.18.0+):** Open Quest, then select
  205. **Better Harness (Beta)** from the left sidebar.
  206. #### Qoder CLI
  207. If Qoder Desktop is installed, Better Harness is already available in Qoder
  208. CLI. No marketplace or plugin installation is required. Start a new Qoder CLI
  209. session in the repository you want to analyze and run the report prompt:
  210. ```text
  211. /better-harness analyze this project's AI coding workflow and generate an evidence-backed report
  212. ```
  213. Only when using Qoder CLI without Qoder Desktop, inspect the current manual
  214. installation disposition before following:
  215. ##### From marketplace
  216. ```bash
  217. # Add the plugin marketplace source
  218. qodercli plugin marketplace add 'https://github.com/QoderAI/better-harness.git'
  219. # Install the plugin
  220. qodercli plugin install better-harness@better-harness
  221. # Check installation
  222. qodercli plugin list
  223. ```
  224. ##### From git
  225. ```bash
  226. # Make sure directory exist
  227. mkdir -p $HOME/.qoder/plugins/marketplaces/
  228. git clone https://github.com/QoderAI/better-harness.git \
  229. $HOME/.qoder/plugins/marketplaces/better-harness --depth 1
  230. qodercli plugin install $HOME/.qoder/plugins/marketplaces/better-harness
  231. ```
  232. Replace `.qoder` to `.qoder-cn` in urls for Qoder CN series.
  233. Then start a new Qoder CLI session before using `/better-harness`.
  234. ### Cursor
  235. The Cursor plugin is not published to the marketplace. The repository carries
  236. the source-local manifest, but the current local Cursor help does not verify the
  237. historical `--plugin-dir` contract. Better Harness therefore reports the
  238. installation plan as unavailable instead of emitting that command:
  239. ```bash
  240. git clone https://github.com/QoderAI/better-harness.git
  241. better-harness plugin plan install --host cursor --surface agent --scope session
  242. ```
  243. Cursor session evidence is supported through workspace-matched transcripts,
  244. metadata, and audit logs. A session that was loaded through a separately
  245. verified native route can be checked with `better-harness plugin verify --host
  246. cursor --surface agent`; partial or unavailable coverage remains explicit.
  247. ### GitHub Copilot
  248. Register this repository as a Copilot plugin marketplace, then install Better
  249. Harness:
  250. ```bash
  251. copilot plugin marketplace add QoderAI/better-harness
  252. copilot plugin install better-harness@better-harness
  253. ```
  254. Verify that the Skill loaded:
  255. ```bash
  256. copilot plugin list
  257. ```
  258. Prefer marketplace installs. Direct repository, URL, and local-path installs are
  259. deprecated in Copilot CLI.
  260. Copilot session evidence is supported through workspace-matched Copilot CLI
  261. transcripts under `~/.copilot/session-state/`. Copilot records no per-response
  262. token usage, and VS Code Copilot Chat has no supported durable transcript; both
  263. remain explicit evidence boundaries.
  264. ### Inspect and plan plugin lifecycle changes (Beta)
  265. The standalone CLI can inspect local Better Harness installation evidence for
  266. every host without contacting a registry or changing host configuration:
  267. ```bash
  268. better-harness plugin status --host all
  269. better-harness doctor --platform all
  270. ```
  271. Build a host-specific install, update, or removal plan before using that host's
  272. native UI or CLI. Plans preserve native steps as typed argv data for deliberate
  273. external execution; the human view does not turn them into shell command
  274. strings, and Better Harness does not execute them:
  275. ```bash
  276. better-harness plugin plan install --host qwen --surface cli --scope user
  277. better-harness plugin verify --host qwen --surface cli
  278. ```
  279. Host differences remain explicit. Qoder Desktop is bundled, Cursor is
  280. session-only while its native command contract is being reconciled, Pi
  281. lifecycle commands without current native evidence remain manual or
  282. unavailable, and WorkBuddy has no managed Better Harness plugin lifecycle
  283. surface.
  284. ### More adapters
  285. Beyond the hosts above, Better Harness also supports Qwen Code, Pi, Kimi Code,
  286. WorkBuddy, and Grok. Their exact install, invocation, and
  287. evidence boundaries live in the docs so this README stays focused:
  288. - **Qwen Code** — [installation guide](docs/docs/installation.mdx#qwen-code)
  289. (`qwen extensions install QoderAI/better-harness`).
  290. - **Pi** — [Host Adapter Matrix](docs/docs/hosts/adapter-matrix.md#pi)
  291. (`pi install <source>` or `pi -e <source>`). The same adapter reads
  292. [Oh My Pi (OMP)](docs/docs/hosts/adapter-matrix.md#oh-my-pi-omp) sessions when
  293. `PI_CODING_AGENT_DIR` points at an OMP agent directory.
  294. - **Kimi Code** — [Host Adapter Matrix](docs/adapters/README.md)
  295. (`.kimi-plugin/plugin.json` plugin install).
  296. - **WorkBuddy** — [Host Adapter Matrix](docs/docs/hosts/adapter-matrix.md#workbuddy).
  297. - **Grok** — [Host Adapter Matrix](docs/docs/hosts/adapter-matrix.md#grok).
  298. Each produces a self-contained `report.html` with paired `report.md` and
  299. `findings.json`; missing or partial session evidence stays explicit.
  300. ## Develop and package from source
  301. Development requires Node.js `>=22.20.0 <25.0.0` and npm
  302. `>=10.9.3 <12.0.0` on Windows, macOS, or Linux.
  303. ```bash
  304. npm ci
  305. npm test
  306. npm run pack:verify
  307. ```
  308. Build the source-local Codex plugin artifact with:
  309. ```bash
  310. node scripts/packaging/build-host-plugin.mjs
  311. ```
  312. The validated artifact is written to `dist/plugins/better-harness`.
  313. From the same source checkout, inspect repository evidence without reading local
  314. sessions:
  315. ```bash
  316. node scripts/better-harness.mjs report --no-sessions
  317. ```
  318. From a source checkout, `npm run preview -- --open` serves a bundled fixture.
  319. Canvas preview requires an installed Qoder runtime, or an explicit
  320. `--sdk-media`/`--sdk-root` path. It listens on `127.0.0.1` by default and is a
  321. local inspection tool, not an authenticated sharing service.
  322. ## Contribute
  323. You do not need to understand the whole runtime to contribute. Start with the
  324. smallest surface that matches the improvement you want to make:
  325. | What you can contribute | Start here | Example contribution |
  326. | --- | --- | --- |
  327. | Workflow guidance and engineering practices | [`skills/`](skills/) or [`references/`](references/) | Add sourced guidance for a language, framework, review pattern, or recurring agent workflow. |
  328. | Evaluation models and executable analysis | [`models/`](models/) or [`scripts/`](scripts/) | Add an evidence-backed evaluation lens, detector, or agent-friendly analysis command with fixtures and tests. |
  329. | Delivery controls and host support | [`hooks/`](hooks/) or the [new Coding Agent guide](docs/adapters/contributing-new-coding-agent.md) | Add a narrow lifecycle check or document and validate evidence support for another Coding Agent host. |
  330. | Reports and visual language | [`templates/reporting/`](templates/reporting/) or [`templates/style/`](templates/style/) | Add a report mode, reusable reporting contract, or directive-only visual style with validation evidence. |
  331. | Examples and operating models | [`case-studies/`](case-studies/) | Share a redacted, evidence-bounded example of how a team applies Agent Work Loop analysis and delivery practices. |
  332. To get started:
  333. 1. Read the [community extension map](docs/community.md) to choose the canonical
  334. owner and understand its contract.
  335. 2. Follow the [contribution guide](CONTRIBUTING.md) to set up the project and
  336. scope the change.
  337. 3. For host support, follow the
  338. [new Coding Agent contribution guide](docs/adapters/contributing-new-coding-agent.md)
  339. and update the [host adapter matrix](docs/adapters/README.md).
  340. 4. Add tests, fixtures, or preview evidence when the contribution changes
  341. runtime behavior or rendered output.
  342. 5. Open a focused pull request that explains what changed, why, and how it was
  343. validated.
  344. Not sure where an idea belongs? [Open an issue](https://github.com/QoderAI/better-harness/issues)
  345. before building a new top-level surface or changing a public report, schema,
  346. packaging, or compatibility contract.
  347. ## License
  348. Better Harness is licensed under the [MIT License](LICENSE).
  349. ---
  350. <p align="center">
  351. If Better Harness helps you improve your agent workflow, consider giving it a ⭐
  352. — it helps others find the project.
  353. </p>

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.