Make complex
systems
make sense.

The explicitly selected 2.0.0-beta.1 release adds synthetic/aggregate calculations and human-review drafts. Fresh npm installation was verified on BuilderBob. Other hosts and live model quality need separate acceptance.

Focused tools cover facilities, ownership, quality, claims, finance, workforce, community health, and research, with source metadata and local caches.

An evidence-bound observatory for authoritative source metadata, access requirements, and possible joins. Eight read-only machine/native WebMCP inspection operations expose published metadata; the research planner remains disabled. The tools acquire no data, perform no analysis, and do not authorize access.

Typed metrics, explicit missingness, reusable sources, hospital rosters, and facility maps. The repository distinguishes MIT analytics and ingestion packages from proprietary application packages.

A ChatGPT Apps SDK app and private Codex plugin using exported public metadata. V1 routes non-PHI administrative work with source context and human-review handoffs.

A bounded-memory Rust updater and resolver. Opaque HSIDs preserve identity through renames, with explicit lifecycle evidence and a deliberately small public identity contract.

A Windows application for importing, searching, filtering, and exporting large hospital pricing files, using Polars, DuckDB, and an optional Rust search extension.

A patient-access assessment practice covering voice assistants, chat agents, websites, and portals, with methodology, deliverables, and an inquiry flow.

Coordinate Grok, Cursor, and Muse across up to eight independent assignments, with isolated workspaces, durable task receipts, and results returned for review.

An experimental coordination and memory system with task identity, project knowledge, a bounded forum, and durable history. The repository describes prototype contracts, not a deployed service.

Codex skills for evidence-backed work loops, backlog planning and cleanup, and technical decision research. The former System Mapper is deprecated.

Mission-oriented skills map repositories, formulate requirements, qualify changes, and expose proof gaps and launch blockers through inspectable local artifacts.

A native Windows utility with a configurable monitor-edge animation and an optional completion island with result and quick reply.

A short-lived local form writes a user-supplied credential to a confirmed file and returns request status to Codex. The destination remains readable to processes with the same OS permissions.

A Codex skill for recruiting the human’s context, preferences, authority, physical action, or verification while keeping workflow coordination with the agent.

162 documented tools connect MCP clients to CATIA Magic / Cameo Systems Modeler for SysML and UML querying, editing, validation, diagrams, and methodology-aware workflows.

Five specialist agents, 20 examples, and mappings across six MBSE tools bring standards context and reviewer preparation into seven supported assistant platforms.

A Rust verification project that combines declarative specifications, systematic state exploration, and WebAssembly execution to produce reproducible discrepancies.

A working vision for Capella/Arcadia integration with engineering rules and composable specialist agents represented as searchable data. This is a concept, not a shipped platform.

A Rust prediction-market system in foundation development, with proposed intelligence, risk, execution, and reconciliation layers. No performance or readiness claim.

Persona-agent research, live market adapters, an MCP surface, and a replayable paper-trading engine. The repository does not expose real-order execution.

The lab’s journal and publishing platform brings together articles, podcasts, strategy insights, a digital bookstore, and institutional access.

News, events, voices, a local-product watch, and a shared feed for humans and agents in the Delaware Valley and Greater Philadelphia.

A focused Navy Yard opportunity navigator with a 30-record launch dataset, official-source links, eligibility and availability context, and verification notes.

Essays, project cases, operating notes, and public artifacts connect the lab’s work to Cole’s healthcare analytics and workflow evaluation practice.