marianne

v0.4.0

Marianne AI Compose plugin for composing scores, authoring configs, submitting scores to the conductor, debugging failed runs, and embedding Marianne behind apps.

CodexClaude Code10 Skills

By EmziLicense: MIT0 GitHub starsUpdated 4 days ago

Directory evidence

Runtimes: Codex and Claude Code
Parsed components: 10 skill or MCP entries
Source updated: Sep 19, 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 marianne 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 marianne@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/Mzzkc/marianne-ai-plugins

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

Plugin files

marianne/

├── .codex-plugin/plugin.json
├── .claude-plugin/plugin.json
├── skills/command/SKILL.md
├── skills/composing/SKILL.md
├── skills/conducting/SKILL.md
├── skills/marianne-embed/SKILL.md
├── skills/marianne-expert/SKILL.md
├── skills/marianne-model-profile-refresh/SKILL.md
├── skills/marianne-search/SKILL.md
├── skills/research/SKILL.md
├── skills/score-authoring/SKILL.md
└── skills/thinking-lab/SKILL.md

Included Skills10

commandskills/command/SKILL.md

Use when submitting, monitoring, debugging, or recovering Marianne score and persistent-agent lifecycle runs. Covers runtime job_id handles, delivery receipts, lifecycle debt, diagnostics, recovery, and anti-patterns that lose work. Do NOT use for writing score configs (use score-authoring instead).

composingskills/composing/SKILL.md

Use when turning a goal into a Marianne score, persistent-agent engagement, multi-stage AI workflow, concert, fan-out, evaluator loop, or release-grade orchestration YAML.

conductingskills/conducting/SKILL.md

Use when directing, steering, supervising, recovering, or judging a Marianne score, persistent-agent lifecycle, concert, fleet, or other asynchronous expert performance toward a shared outcome.

marianne-embedskills/marianne-embed/SKILL.md

Use when embedding Marianne behind another app, local dashboard, web service, automation wrapper, or agent-facing tool. Covers daemon IPC bridges, hidden score submission and runtime job_id monitoring, score validation, runtime variables, conductor lifecycle guardrails, and building a product-facing abstraction over Marianne without exposing Marianne internals. Do not use for authoring score YAML from scratch or ordinary CLI run debugging unless the task is specifically about wrapping those operations.

marianne-expertskills/marianne-expert/SKILL.md

Use when Marianne composing, persistent-agent construction or lifecycle, operating, debugging, architecture, source development, embedding, runtime commissioning, or current-versus-pinned evidence reconciliation is required.

marianne-model-profile-refreshskills/marianne-model-profile-refresh/SKILL.md

Use when users ask to "update Marianne models", "refresh instrument profiles", "upgrade musician profiles", "audit stale model IDs", or "run the Marianne model updater".

marianne-searchskills/marianne-search/SKILL.md

Compatibility entry for the sibling research skill's caller-bound adaptive search score.

researchskills/research/SKILL.md

Research existing components or obtain independent supplied-context review, with caller-owned synthesis when using the lab.

score-authoringskills/score-authoring/SKILL.md

Use when writing, reviewing, or fixing Marianne score YAML configs, including persistent-agent identity, memory, technique, cadenza, and lifecycle attachment. Do NOT use for running/debugging jobs (use command instead).

thinking-labskills/thinking-lab/SKILL.md

Compatibility entry for the sibling research bundle's independent supplied-context review score.

Plugin manifests2

marianne/.codex-plugin/plugin.json

{
  "name": "marianne",
  "version": "0.4.0",
  "description": "Marianne AI Compose plugin for composing scores, authoring configs, submitting scores to the conductor, debugging failed runs, and embedding Marianne behind apps.",
  "author": {
    "name": "Emzi",
    "url": "https://github.com/Mzzkc"
  },
  "repository": "https://github.com/Mzzkc/marianne-ai-compose",
  "license": "MIT",
  "keywords": [
    "marianne",
    "orchestration",
    "scores",
    "composition",
    "ai-compose"
  ],
  "skills": "./skills/",
  "interface": {
    "displayName": "Marianne",
    "shortDescription": "Compose, run, debug, and embed Marianne scores",
    "longDescription": "Tools and skills for building with Marianne AI Compose, including score authoring, conductor operations for submitted scores, thinking-lab workflows, and embedding Marianne behind local apps or agent-facing services.",
    "developerName": "Emzi",
    "category": "Productivity",
    "capabilities": [
      "Write",
      "Interactive"
    ],
    "defaultPrompt": [
      "Use Marianne to author and validate a score.",
      "Debug a failed Marianne score run.",
      "Embed Marianne behind a local app."
    ]
  }
}

marianne/.claude-plugin/plugin.json

{
  "name": "marianne",
  "version": "0.4.0",
  "description": "Marianne AI Compose plugin — compose scores, author configs, submit scores to the conductor, and debug failed runs.",
  "author": {
    "name": "Emzi",
    "url": "https://github.com/Mzzkc"
  },
  "repository": "https://github.com/Mzzkc/marianne-ai-compose",
  "license": "MIT",
  "keywords": [
    "marianne",
    "orchestration",
    "scores",
    "composition",
    "ai-compose"
  ]
}

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.

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