← Back to Skills Marketplace
nutt-adam

Tutti

by nutt-adam · GitHub ↗ · v1.1.0 · MIT-0
cross-platform ✓ Security Clean
327
Downloads
0
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install tutti
Description
Orchestrate multiple AI coding agents (Claude Code, Codex, Aider) from a single config — launch teams, run workflows, track capacity, and manage handoffs.
Usage Guidance
This wrapper is coherent: it delegates work to the local 'tt' CLI and reads .tutti/state/*.json for status. Before using, verify the provenance of the 'tt' binary (https://github.com/nutthouse/tutti or another trusted source) because the skill runs whatever 'tt' does (and workflows/steps can execute shell commands and inject workspace files). Only run this skill in repositories you trust, and confirm you are comfortable with the behavior of 'tt' (it is the component that may reach network services or require API keys). Note: TUTTI_BIN can override the command; ensure that env var or overridden binary is trusted.
Capability Analysis
Type: OpenClaw Skill Name: tutti Version: 1.1.0 The 'tutti' skill bundle is a functional wrapper for the Tutti (tt) CLI tool, designed to orchestrate multiple AI coding agents. The Python script `tutti_openclaw.py` provides a structured JSON interface for lifecycle management, workflow execution, and git operations. The logic is transparent, lacks obfuscation, and includes security-conscious instructions in `SKILL.md`, such as escalating authentication failures to the user rather than retrying. No evidence of data exfiltration or malicious intent was found.
Capability Assessment
Purpose & Capability
The name/description (multi-agent orchestration) align with the actual behavior: the skill invokes a local 'tt' CLI, reads .tutti/state/*.json for status, and provides wrapper actions for launch, workflow, handoff, git land, etc. Required binaries (tt, tmux, python3) are appropriate for this functionality.
Instruction Scope
SKILL.md and the wrapper script limit runtime actions to calling the configured 'tt' CLI, reading .tutti/state/*.json, peeking agent output, and copying/injecting workspace-relative files for handoffs. There are no instructions to read unrelated system files or environment secrets.
Install Mechanism
There is no install spec (instruction-only skill with one wrapper script). Nothing is downloaded or extracted by the skill itself. The only runtime dependency that matters is the external 'tt' CLI (the skill documents its upstream repository).
Credentials
The skill declares no required environment variables or credentials. The wrapper optionally respects TUTTI_BIN if set, which is reasonable. No unrelated secrets or config paths are requested.
Persistence & Privilege
The skill does not request always: true, does not modify other skills' configs, and does not claim persistent system-level privileges. It runs on-demand and invokes the local 'tt' tool as needed.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install tutti
  3. After installation, invoke the skill by name or use /tutti
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.1.0
Add send_prompt, land_agent actions and workflow step type docs
v1.0.0
Initial release
Metadata
Slug tutti
Version 1.1.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 2
Frequently Asked Questions

What is Tutti?

Orchestrate multiple AI coding agents (Claude Code, Codex, Aider) from a single config — launch teams, run workflows, track capacity, and manage handoffs. It is an AI Agent Skill for Claude Code / OpenClaw, with 327 downloads so far.

How do I install Tutti?

Run "/install tutti" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Tutti free?

Yes, Tutti is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Tutti support?

Tutti is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Tutti?

It is built and maintained by nutt-adam (@nutt-adam); the current version is v1.1.0.

💬 Comments