What is harness-evolver-marketplace?

A plugin marketplace listing for harness-evolver, a tool that changes language-model agent code through multiple proposing agents, LangSmith experiments, and Git worktrees. LangSmith is a service for tracing and evaluating language-model applications; Git worktrees are separate working directories for different branches.

Use it to evolve agent code, run LangSmith experiments, and manage competing code changes in separate Git worktrees.

It provides a recorded way to run and compare proposed changes to agent code during optimization.

harness-evolver-marketplace — Plugin by raphaelchristi

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Claude Code

/plugin marketplace add raphaelchristi/harness-evolver

agentmods

npx agentmods add plugins/raphaelchristi/harness-evolver/marketplace

Clone the repo

git clone --depth 1 https://github.com/raphaelchristi/harness-evolver

Made for: Claude Code.

Per session not measured What this adds to a session before it is invoked.

When invoked not measured Not applicable: nothing here is loaded into a session.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Security

Grade A, and why

harness-evolver-marketplace scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude-plugin/marketplace.json · 26 lines

What it actually says

{
  "name": "harness-evolver-marketplace",
  "owner": {
    "name": "Raphael Valdetaro"
  },
  "metadata": {
    "description": "LangSmith-native autonomous agent optimization plugin"
  },
  "plugins": [
    {
      "name": "harness-evolver",
      "source": "./",
      "description": "Evolves LLM agent code using multi-agent proposers, LangSmith experiments, and git worktrees",
      "version": "6.1.0",
      "author": {
        "name": "Raphael Valdetaro"
      },
      "homepage": "https://github.com/raphaelchristi/harness-evolver",
      "repository": "https://github.com/raphaelchristi/harness-evolver",
      "license": "MIT",
      "keywords": ["langsmith", "optimization", "evolution", "llm", "agent"],
      "category": "development"
    }
  ]
}

Files

What ships with it

1 file beside marketplace.json in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

plugin.json 564 B

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

2d ago First seen · 26 lines scan A 894af21166aa

Subscribe to this mod's changes

harness-evolver-marketplace is a plugin published in the GitHub repository raphaelchristi/harness-evolver (49 stars, last pushed 4mo ago), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Agent Evolution Data and AI Git and workflow claude-code-plugin codex-skills harness-engineering meta-harness

better-harness

Build an AI-ready engineering system for safe coding-agent delivery and continuous software improvement.

QoderAI/better-harness · not measured

claude

Connect Claude Code to your local OpenLoomi runtime. One-command install, mirror Claude's lifecycle onto the Loomi Pet, and auto-archive every Stop into your OpenLoomi memory. Talks to your local OpenLoomi over HTTP and the OpenLoomi helper CLI. AI provider config is owned by OpenLoomi itself — this plugin never reads.

melandlabs/openloomi · not measured

heliohq

Helio plugin registry.

heliohq/ship · not measured

workflow

Developer-workflow commands and skills — sandbox-safe Conventional commits with a commit.sh handoff, immutable code-scanning triage reports with SHA-pinned permalinks, and generation of evolve evaluation suites (triggers + behavioral evals) for agent skills.

bitwise-media-group/skills · not measured

terraform

Terraform style conventions, module scaffolding, and fmt/validate/tflint workflows for authoring reusable modules.