Tools All levels 2 minute read Updated 2026-06-29 UTC

Browser-Local Tools

Local tools for model-breeding education: fitness scorecards, population simulation, lineage viewing, router policy, merge planning, adapter stacks, and release packets.

Research statusBrowser-local decision support Publication statePublished Reviewed byMichael Kappel Source reports8

Answer first

The tools section turns ModelBreeder theory into browser-local worksheets. Use the scorecards, simulators, router lab, lineage viewer, and release packet builder to reason about useful descendants before implementing a real model-breeding runtime.

Tools for constructive model-breeding decisions

The tools make the theory visible. They run in the browser, require no database, send no data to a remote form handler, and are designed as teaching surfaces that can become real project worksheets.

ToolWhat it teachesOpen it
Local AI Adoption PlannerScore a workflow for local-first, hybrid, or lab-first adoption.Open
Fitness Scorecard CalculatorCompare a descendant with champion, specialist, challenger, and no-op choices.Open
Local AI Opportunity ScorecardPick the first private, repeated, local-first workflow for a useful specialist.Open
Population SimulatorMutation, selection, carrying capacity, diversity, and retirement.Open
Evolution DashboardPopulation state, novelty spread, role assignment, and next action.Open
Lineage DAG ViewerParentage, operators, evidence, lifecycle state, and rollback.Open
Router Policy LabLocal specialist, adapter stack, cascade, ensemble, review, and no-op.Open
Release Packet BuilderCopyable evidence packet for shadow, canary, promotion, rollback, or archive.Open
Merge Recipe SketchpadParent selection, operators, coefficients, and equal-budget evaluation.Open
Adapter Stack PlannerBase model, adapter order, compatibility, and local runtime fit.Open
Model Ecology Glossary ToolFast lookup of core vocabulary and where to go next.Open

Use the outputs as evidence starters

A scorecard or simulator result is not a final truth. It is a structured starting point for an evidence packet, architecture decision record, or lab note.

pseudocode
PROCEDURE use_browser_tool(tool, project_context)
    assumptions <- RECORD_ASSUMPTIONS(project_context)
    result <- tool.RUN_LOCALLY(project_context)
    next_step <- SELECT_NEXT_EVIDENCE_ACTION(result)
    RETURN CREATE_LAB_NOTE(assumptions, result, next_step)
END PROCEDURE

Local AI adoption tools

Source reports used for this guide

These reports are preserved verbatim in the site archive. The guide above is an editorial synthesis and may narrow, qualify, or reorganize claims from the source material.