Agent Plugins Marketplace
← All plugins

goalspec

v0.44.9

Make any agent write its own grounded goal-spec, verify it with an independent adversary, and close through a fail-open completion gate. Hardens the built-in /goal.

Claude Code1 Skill

By vzertLicense: MIT2 GitHub starsUpdated 6 days ago

Directory evidence

Runtimes
Claude Code
Parsed components
1 skill or MCP entry
Source updated
Sep 18, 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 goalspec for Claude Code

Installs for the current user
claude plugin marketplace add IchenDEV/agent-plugin-mkt
claude plugin marketplace update agent-plugin-marketplace
claude plugin install goalspec@agent-plugin-marketplace

Paste and run these commands in a terminal with Claude Code. 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/vzert/goal-forge

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

Plugin files

plugins/goalspec/
├── .claude-plugin/plugin.json
└── skills/adversary/SKILL.md

Included Skills1

adversaryskills/adversary/SKILL.md

Standalone adversarial verification of the current conversation's claimed outcome, without running the full goalspec loop. Manual-first — invoke as /goalspec:adversary when the user explicitly asks for an independent adversary to try to BREAK a specific claimed result ("verify this with the adversary", "red-team this outcome", "have an independent verifier try to break it"). Do NOT auto-trigger on generic review requests ("check this", "review my work", "does this look right") — those are an ordinary review, or the full /goalspec method. Builds the pointer payload, routes through the configured adversary backend, and quotes the verdict verbatim; a bare hold is UNVERIFIED, never a pass.

Plugin manifests1

plugins/goalspec/.claude-plugin/plugin.json
{
  "name": "goalspec",
  "version": "0.44.9",
  "description": "Make any agent write its own grounded goal-spec, verify it with an independent adversary, and close through a fail-open completion gate. Hardens the built-in /goal.",
  "author": {
    "name": "vzert"
  },
  "repository": "https://github.com/vzert/goal-forge",
  "license": "MIT",
  "keywords": [
    "goal",
    "spec",
    "adversary",
    "verification",
    "constitution",
    "autonomy"
  ]
}

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.

[goalspec on Agent Plugins Marketplace](https://pluginsmp.com/plugins/goalspec)