Skip to content
Tools/Agent Orchestrators
Agent OrchestratorsCommunityOpen source

tutti

Multi-agent orchestration CLI driven by a tutti.toml — chains prompt steps, shell commands, and agents into repeatable pipelines with git worktree isolation.

Visit source
Use casesMulti-agent orchestratorsWorktree tools
Reviewed 2026-07-21
Community project documentation
Primary source reviewed

Source type: open-source. License: MIT. The linked source supports the visible capability notes but does not prove production suitability.

Open evidence source
Inference or recommendation
Verify before production

Confirm current maintenance, permissions, data handling, license obligations, failure behavior, and a human approval gate in your own environment.

Key capabilities

Repeatable pipelines defined in tutti.toml (prompt steps + shell commands + agent coordination)
Isolates each agent in its own git worktree to prevent merge conflicts
Typed artifact flow: capture via artifact_glob / artifact_name, inject downstream with inject_files
License: MIT

Practical fit

tutti fits repeatable pipelines where prompts, shell commands, agents, and file artifacts need an explicit order. Its typed artifact handoff is useful when the output of one step must become controlled input to another.

Configuration does not make a pipeline safe by itself. Shell steps and injected files still need scope controls, input validation, and a clear owner for conflicts or partial output.

How to evaluate it

Can every pipeline step declare a bounded input and output artifact?
What happens when one worktree or command fails halfway through?
Are injected files reviewed before they become instructions for another agent?

Example decision

Build a two-step research pipeline in a disposable repository: one agent writes a sourced inventory artifact, and a second agent checks it against a rubric without editing the first agent's worktree.

Verify before you adopt it

Check current tutti.toml syntax and artifact behavior in the repository.
Review every shell command and the MIT license.
Test cleanup and retry behavior after an intentional failure.

Source review note

We reviewed the primary source on July 21, 2026. Product behavior, pricing, licenses, and maintenance status can change, so check the linked source before adopting it.

Best for

Config-driven multi-agent pipelines
Worktree isolation
Typed artifact hand-off between steps

Not for

One-off prompts

Risk notes

Community project — evaluate before production use.
This is a community project. Verify maturity, licensing, permissions, and safety behavior before relying on it in production.