this one needs WebGL
The explorer is a 3D graph you orbit and descend into, and this browser could not start a WebGL context. Everything in it is on the main page, in text.
← read it flat insteadA navigable graph of my work, structured the way chrono-mind navigates vectors: sparse overview on top, full detail below.
A coordination runtime for agents running as separate processes: rooms, atomic task claims, shared state, presence, approval gates and an audit log.
Permissions, claims and audit run inside one transactional reducer on SpacetimeDB, so two agents can never both own a task. Ships an MCP server, LangGraph and CrewAI adapters, and phase-based blueprints for multi-agent workflows.
Sixteen agents see the same task at the same moment. Release it and watch who gets it — then switch to a naive read-then-write claim and watch it go wrong.
Rust, Python, SpacetimeDB, MCP, LangGraph
A marketing agency run by agents, where anything that spends money or publishes stops for a named human.
Approval gates are LangGraph interrupts that survive a restart. A rejection goes back to the agent as the tool result it reads next. Every approval shows what was checked — and what nobody checked. Committed evals fail CI when the reviewer agent gets lenient.
One campaign, start to finish. Step through it: the spend is proposed, frozen, rejected, revised, and only then allowed to happen.
Python, deepagents, LangGraph, evals
A StarCraft II agent that treats attention as a budget, using Bayesian surprise to decide which moments are worth thinking about.
Perception, strategy and micro run as separate stages; an information-gain scheduler skips frames that would not change a decision. Every game is logged as JSONL, losses included.
Attention as a budget. Most frames change nothing; the agent spends thought only where the game surprises it.
Python, Bayesian surprise, python-sc2
An open-source, schema-agnostic backend for multi-agent systems: any agent can log interactions under schemas registered at runtime, and anyone can question the resulting graph in plain English.
FastAPI and ArangoDB underneath; JSON schemas are registered, versioned and enforced while the system runs. Questions become AQL through local Qwen3 models on Ollama — then every generated query passes a safety gate: writes blocked, collections whitelisted, results capped. The LLM subsystem was rewritten from 5,107 to 815 lines and got more reliable doing it. A React front end turns answers into charts and live network views.
Agent interactions land in a graph store under schemas registered at runtime. Natural language becomes AQL — but only through a safety gate.
Python, FastAPI, ArangoDB, Ollama + Qwen3, React, WebSocket
BackendFrontendA research agent that reads papers, extracts claims, and commits only the ones a different model family can find in the source.
It refuses to start if extractor and verifier share a model family. Unsupported claims are kept, marked uncommitted, for audit. Surviving claims form a graph that is mined for gaps and argued over by a proposer and a cross-family adversary.
Run the gate with one verifier, then with two from different families, and watch which fabrications survive the first pass.
Python, pgvector, cross-family verification
CodeA theorem-proving agent loop where the Lean 4 kernel is the only judge.
Statements are formalised, compiled and repaired with the compiler error fed back. An integrity gate rejects sorry, admit and untrusted axioms before anything counts. Verified proofs are stored as content-addressed nodes in a shared proof graph.
A proof loop with one judge. The model may propose anything; only what the Lean kernel accepts, and the integrity gate lets through, counts.
Python, Lean 4, multi-agent, eval harness
Breaking-change detection for the APIs and MCP servers agents call, so a tool changing under an agent fails a build instead of a task.
Fetch, normalise, diff against the last snapshot, and classify each change as strict-breaking, agent-breaking or safe — an LLM classifier with deterministic rules underneath. Never serves stale data as current. Exposed as a feed, a REST API, an MCP server and a GitHub Action.
Tools change under agents. This finds the changes a compiler would never flag but an agent would trip on.
TypeScript, MCP, GitHub Actions
FeedActionA live learning platform that turns a job posting into a study route, then has an AI tutor judge real work — with calibration suites that fail when the judge gets flattering.
Competencies are matched against 75 module contracts and closed over prerequisites; the route states plainly what it cannot teach. The tutor scores on three dimensions and asks one question only the author could answer. Both the matcher and the tutor have calibration suites that make real model calls and exit non-zero.
From a job posting to proven work. Two places where a model judges — and both are calibrated, so flattery fails a test.
Python, FastAPI, Postgres, Astro
CodeLiveA lock-free vector index with time-aware memory for agents: wait-free reads, lock-free writes, recency and importance in the ranking.
Correctness is checked, not assumed: loom model checking, Miri, ThreadSanitizer, a differential oracle and fuzzing all run in CI. Under concurrent writes it keeps 65% of read throughput where a single RwLock keeps 1%.
A real query, running. It enters at the sparse top layer and descends greedily — touching a handful of nodes instead of all of them. Toggle decay to let recency compete with similarity.
Rust, HNSW, loom, Miri, PyO3
CodeA long-running companion whose memory is split into faculties: a semantic graph, time-decayed episodes, affect, and a council for hard questions.
Facts are extracted by one model family and verified by another before they are committed. Recall blends similarity with a ~21-day half-life and importance. Background tasks run plan, execute, verify, escalate — with a watchdog behind them.
One message through a memory made of faculties. The reply is fast; what gets remembered is slower, and checked by a second model family first.
Python, knowledge graph, cross-family verification
A behavioural layer that turns a person’s state into constraints any LLM can follow — context, never actions.
23 interpretable axes; what a person says about themselves always overrides what is inferred. Explicit non-goals: no action execution, no persuasion optimisation, no hidden optimisers.
23 interpretable axes collapse into a state vector, which a rule-based translator turns into prompt constraints. It emits context, never actions.
Rust, Python, model-agnostic
Code