Skip to main content

Open-source tools for accountable AI work

Delegate with clear scope.
Keep the evidence.

Attune Harness runs agent tasks, checks results independently against criteria you supply, and keeps the output, check results and errors for review.

Collaborate

Discuss a pilotSponsor open-source milestones

Attune Harness

A task you can inspect.

  1. Agree the scopeGoal, limits and acceptance criteria
  2. Authorize the workPermitted actions and recorded output
  3. Check the resultIndependent checks, outcome and errors
Illustrative workflow. A real receipt records what ran and what the checks established.

Help shape what comes next

Three ways to take part.

Proposed paths for contributors, potential pilot customers and open-source sponsors.

Start by building together

Collaborate

Help build and test tools for accountable agent work. Bring a reproducible problem, a design question or a contribution.

Explore collaboration

A starting conversation could identify a useful contribution, its scope and how to check it. Documentation, usability and real workflow evidence matter alongside code.

Start a conversation

Explore a possible customer engagement

Discuss a pilot

Start with one bounded workflow and a result we can check. Assess the fit and define what a possible pilot would need to demonstrate.

Explore a pilot

Possible starting points include reviewing a generated document against project evidence or checking a scoped agent-assisted change. Scope, responsibilities and commercial terms would need separate agreement.

Start a conversation

The lead project · Attune Harness

Make the work inspectable.

Plan, build, review, fix and test with an explicit scope and retained evidence.

Trustworthy delegation starts with knowing what was authorized, what happened and how the result was checked. Harness records these so you can judge the outcome.

Authorization has limits.

Accepting intent and authorizing execution are separate decisions. Paid calls need their own authorization.

Evidence has limits too.

A passing check establishes what it tested. It does not guarantee every output is correct.

Published: Attune Harness 1.3.0. Package and release guide

Browser forms guide intake and intent approval. Each workflow retains its own execution permissions and checks.

Read the evidence and limits

A passing check establishes what it tested. Model adapters and platform support have qualification limits; native MCP panel delivery remains unverified. Review the qualification guide.

The wider portfolio

Memory. Decisions. Sources.

Related projects with distinct roles. Harness leads; the existing tools remain available.

Project memory and workflows

Attune AI

v16.4.0

Keep project decisions and lessons available across sessions, with workflows for AI coding agents.

Explore Attune AI

Structured questions and decisions

Attune Forms

Make questions and choices easier to review, with structured answers validated on return.

Explore Attune Forms

Retrieval with citations

Attune RAG

Retrieve relevant project sources and return citation records for the next step.

Explore Attune RAG

Built by Patrick Roebuck

Developed in the open.

The direction is practical: make agent work easier to authorize, inspect and evaluate. The next step is to find useful problems to work on together.