capability-framework

Public evidence package for the Instats report on responsible AI use in PhD-level research and research training. A global capability framework for university research leadership across 38 top-tier doctoral universities and 14 national research funders.

Agent Capability Standard is an open specification for composable AI agent capabilities. It defines 36 atomic capabilities across 9 cognitive layers, a type-safe workflow DSL, and grounded world modeling with trust-aware conflict resolution. Built on the Grounded Agency philosophy, it makes agent reliability structural—not optional.

Historical Reference Application #003 for Agnara. A Python 3.14 project demonstrating execution governance, principals, scopes, operational risk, effects, policy evaluation, confirmation evidence, authorization boundaries and canonical execution outcomes using Agnara 0.1.0a2.

Historical Reference Application #001 for Agnara. A minimal Python 3.14 project demonstrating capabilities, schema inference, execution plans, canonical invocation, native validation, metadata, scopes, risk, and Success/Failure using the public Agnara package from PyPI.

IOWAP documentation — setup, concepts, API reference, node guides

Meta-repo for IOWAP — the distributed node capability framework. Story, architecture, and links to the other repos.

IOWAP relay server — API, scheduler, auth, database, and dashboard for node orchestration.

IOWAP storage node — file storage, retrieval, and bridging for the IOWAP ecosystem.

Historical Reference Application #002 for Agnara. A Python 3.14 project demonstrating dependency injection, providers, DIRegistry, DIContainer, compiled dependency graphs, protected runtime parameters, caller-owned inputs, dependency resolution and execution plans using Agnara 0.1.0a2.

IOWAP node framework — daemon, CLI, capability management, and handler runner. Build and register your own node.