npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

@rulebased/harness

v1.4.7

Published

Harness engineering tools for AI agents — audit, init, recommend, eval-log

Vulnerabilities

Powered byPowered by Snyk
  • 0High
  • 0Medium
  • 0Low

Links

Git

Maintainers

jungyounjungyoun

Keywords

Readme

@rulebased/harness

A harness-building tool for AI agents. Assess how well your project's harness engineering is set up, get recommendations for missing elements, and auto-generate them.

Harness Engineering = A system design approach that constrains agent behavior (Constraints), provides context (Context), and evaluates results (Eval). Based on practices from OpenAI, Anthropic, GitHub, Stripe, Martin Fowler, and more.

Installation

Claude Code Plugin

/plugin marketplace add rulebased-io/claude-plugin
/plugin install rulebased-harness@rulebased

CLI (npm)

npx @rulebased/harness audit
npx @rulebased/harness score
npx @rulebased/harness init
npx @rulebased/harness recommend
npx @rulebased/harness eval-log

skills.sh

npx skills add rulebased-io/claude-plugin --skill harness-audit

Skills

After installation, the following slash commands are available:

/rulebased-harness:audit      # Audit harness coverage (36 checks, 0-100 score)
/rulebased-harness:score      # Per-category detailed score report
/rulebased-harness:init       # Initialize harness structure
/rulebased-harness:recommend  # Recommend missing harness elements
/rulebased-harness:eval-log   # Evaluate conversation log compliance

audit

Checks 36 items derived from industry best practices (OpenAI, Anthropic, GitHub, Stripe, Google, Martin Fowler, Vercel, and others) and assigns a score from 0 to 100.

| Category | Items Checked | |----------|---------------| | Context Engineering | AGENTS.md, ARCHITECTURE.md, subdirectory AGENTS.md, docs/, CLAUDE.md | | Bootstrap & Task Entry | One-command setup, build/test/lint commands, lockfile | | Constraints & Enforcement | Linter, formatter, pre-commit hooks, TypeScript strict | | Eval & CI | CI pipeline, CI lint step, test suite, eval dataset | | Entropy Management | Tech debt tracker, in-repo documentation | | Safety & Secrets | .gitignore, .env blocking, security docs | | Knowledge Management | ADRs, README | | Workflow | specs/, tasks/ directories |

score

Per-category detailed score report with pass/fail per check.

init

Initializes the harness structure: AGENTS.md, CLAUDE.md, .harness.json, specs/, tasks/, .gitignore.

Presets: standard (all 36 checks) or minimal (AGENTS.md + build commands).

recommend

Recommends missing harness elements by priority and can auto-generate them.

eval-log

Evaluates a Claude Code conversation transcript against harness compliance. Auto-triggers via Stop hook when a session with 10+ agent turns ends.

Scoring

| Severity | Weight | Examples | |----------|--------|----------| | Critical | 3 | AGENTS.md, build/test/lint commands, CI pipeline, .gitignore | | Important | 2 | Architecture, linter, formatter, pre-commit, lockfile, ADRs | | Nice-to-have | 1 | Workflow directories, eval dataset, security docs |

Grades: A (90+), B (75+), C (60+), D (40+), F

License

MIT