Agent Plugins Marketplace
← All plugins

behavior-diff

v0.4.2

Test whether an uncommitted CLAUDE.md/AGENTS.md/skill edit actually changes agent behavior: isolated before/after trials (claude, codex, pi, or omp), deterministic trace grading, a decision diff, an HTML evidence report, and a retro skill that feeds lessons back.

CodexClaude Code3 Skills

By Kent Huang1 GitHub starsUpdated last week

Directory evidence

Runtimes
Codex and Claude Code
Parsed components
3 skill or MCP entries
Source updated
Oct 7, 2026
Manifest status
Canonical path parsed

The directory validates manifest shape and source location. It does not execute the plugin or provide a security endorsement. Review the indexing methodology →

Install behavior-diff for Codex and Claude Code

Installs for the current user
codex plugin marketplace add IchenDEV/agent-plugin-mkt
codex plugin marketplace upgrade agent-plugin-marketplace
codex plugin add behavior-diff@agent-plugin-marketplace

Paste and run these commands in a terminal with Codex. They add and refresh the PluginsMP catalog, then install this plugin.

The installer fetches third-party code from the source repository shown on this page. This directory validates manifest structure and source location, but does not perform a security audit; review the manifest, components, and source before installing.

Get the source manually
git clone https://github.com/spacedock-dev/behavior-diff

Clone the source repository, then follow its setup instructions to add the plugin to a compatible client. The plugin root is plugin/.

Plugin files

plugin/
├── .codex-plugin/plugin.json
├── .claude-plugin/plugin.json
├── skills/behavior-diff/SKILL.md
├── skills/behavior-diff-live/SKILL.md
└── skills/behavior-diff-retro/SKILL.md

Included Skills3

behavior-diffskills/behavior-diff/SKILL.md

Compare agent behavior before/after an uncommitted change to CLAUDE.md, AGENTS.md, or a skill file. Use for "behavior diff", "test my CLAUDE.md change", "did my rule change the agent's behavior", "before/after check on my rule edit".

behavior-diff-liveskills/behavior-diff-live/SKILL.md

Run a before/after behavior diff inside the current session using subagents — one trial per variant, the main agent prepares the decision-moment scenario and watches both runs. Use for "live behavior diff", "behavior diff with subagents", "quick behavior diff", or experiment runs where the user wants to adjust the scenario and see progress.

behavior-diff-retroskills/behavior-diff-retro/SKILL.md

Write a retro for a finished behavior-diff or behavior-diff-live run — where the time went, what broke, what the evidence can and cannot support — and feed the lessons back into Behavior Diff. Use for "behavior diff retro", "retro this run", "retro the experiment", after any behavior-diff run finishes.

Plugin manifests2

plugin/.codex-plugin/plugin.json
{
  "name": "behavior-diff",
  "description": "Test whether an uncommitted CLAUDE.md/AGENTS.md/skill edit actually changes agent behavior: isolated before/after trials (claude, codex, pi, or omp), deterministic trace grading, a decision diff, an HTML evidence report, and a retro skill that feeds lessons back.",
  "version": "0.4.2",
  "author": {
    "name": "Kent Huang"
  }
}
plugin/.claude-plugin/plugin.json
{
  "name": "behavior-diff",
  "description": "Test whether an uncommitted CLAUDE.md/AGENTS.md/skill edit actually changes agent behavior: isolated before/after trials (claude, codex, pi, or omp), deterministic trace grading, a decision diff, an HTML evidence report, and a retro skill that feeds lessons back.",
  "version": "0.4.2",
  "author": {
    "name": "Kent Huang"
  }
}

For maintainers

If you maintain this plugin, link to this source-backed listing from your README so users can review its manifest and indexed components.

[behavior-diff on Agent Plugins Marketplace](https://pluginsmp.com/plugins/behavior-diff)