Skip to content

Capability Lab Public Distribution

Capability Lab is experimental research software for evidence-grounded human capability modeling, governed evaluation, current-state selection, advisory progression, and safe external observation.

This repository is the project's current public source-available distribution.

Public lineage

The public Git history intentionally begins from a clean source-available snapshot.

It does not claim to reproduce the complete internal research or development chronology that preceded this public lineage.

Some architecture, reference, legacy, test, or audit documents in the snapshot retain PR- or milestone-oriented names because those names are part of the technical provenance of the work. They are documentation provenance, not published Git ancestry and not a separate license grant.

PUBLIC GIT HISTORY
!= COMPLETE RESEARCH CHRONOLOGY

HISTORICAL DOCUMENT REFERENCES
!= PUBLISHED COMMIT ANCESTRY

PUBLICATION-1 release baseline

The frozen initial public root for this lineage is:

febe79f9630858c2e01e3ed57ae1bfd7736227ba
Initial public source-available release

PUBLICATION-1 does not rewrite that root. It establishes the first ordinary post-root release baseline and makes the lineage boundary executable in hosted CI.

The public-lineage gate requires:

  • exactly one reachable Git root from the proposed public HEAD;
  • that root to equal the frozen initial public root above;
  • the frozen root to have zero parents;
  • the frozen root author and committer metadata to use GitHub's noreply identity;
  • owner-authored commits after the root to use GitHub's noreply identity.
FROZEN PUBLIC ROOT
!= MUTABLE RELEASE METADATA

ONE CLEAN PUBLIC ROOT
!= IMPORTED HISTORICAL GIT ANCESTRY

This gate is a repository-integrity contract. It does not replace the software license, establish ownership by itself, or rewrite rights granted for earlier distributed copies.

Current checkpoint

The current distribution represents a stable research-subsystem checkpoint after the first major architecture cycle.

Future architecture work is demand-driven. Publication of the current snapshot does not pre-authorize new inference semantics, automatic capability updates, progression authority, permission semantics, professional authority, or an automatic closed loop.

PRODUCT / READ PROJECTION
!= CURRENT-STATE SELECTION AUTHORITY
!= PROGRESSION AUTHORITY
!= CAPABILITY UPDATE AUTHORITY
!= AUTOMATIC CLOSED LOOP

Privacy boundary

A public Capability Lab distribution contains software, documentation, tests, and synthetic research artifacts. It is not a public capability database about real people.

Person-scoped evidence, capability state, participant workspaces, raw captures, private histories, credentials, and similar sensitive records remain private by default.

Publishing Capability Lab source code does not change the privacy or human-agency boundaries defined by the architecture.

Licensing

The current public distribution uses the PolyForm Noncommercial License 1.0.0 (PolyForm-Noncommercial-1.0.0). See LICENSE.

SOURCE-AVAILABLE
!= OSI OPEN SOURCE

PUBLIC SOURCE ACCESS
!= COMMERCIAL USE PERMISSION

Commercial uses that require rights beyond the public license require separate written commercial licensing. See COMMERCIAL-LICENSING.md.

Earlier versions of Capability Lab were previously distributed under Apache-2.0. Rights already granted for those earlier copies remain in force for those copies; the current license does not retroactively rewrite them. See LICENSE-HISTORY.md.

Nothing in this publication record narrows fair use or other rights provided by applicable law.

Contributions

Research discussion, bug reports, architecture ideas, documentation feedback, and conceptual adversarial cases are welcome.

Until a dedicated contributor-rights agreement or equivalent process is published, substantive third-party authored material is not accepted for inclusion. This preserves the rights chain required for separate commercial licensing. See CONTRIBUTING.md.

Development status

Capability Lab is published as a stable research subsystem rather than an automatically expanding product roadmap.

New work should be driven by concrete consumer needs or research questions and must preserve the project's separation between evidence, inference, state, advice, and authority.