00 — vibes
42% complete — trust me bro

we build AI agent infrastructure and publish the math behind it

01 — METHODOLOGY

we're very serious about not being serious

we think building things is about to get extremely weird. like, 'a thousand robots doing incredible* things simultaneously' weird.

so we're building the infrastructure for that. agent-native, specification-first, and deeply suspicious of anything that only works because a human is watching.

software engineering is just the start.

*incredible only if us humans get this right

  • Deterministic Core Logic
  • Stochastic Surface Layers
  • Agent-Native Architecture
00

controlled chaos

algorithms that curate the noise so you don't have to. every interaction is intentional, even when it looks accidental.

01

vibes-responsive UI

interfaces that read the room. minimal when you're focused, detailed when you're exploring, absent when you're on a roll.

02

humans in the loop (reluctantly)

we'd automate everything if we could. but human intuition is annoyingly irreplaceable. for now.

02 — PRODUCT

modyl

presence, not prompts.

check it out

real-time meeting infrastructure where AI agents join as actual participants — with cameras, voices, screens, and interactive surfaces. not another chatbot. the room is the interface.

specialist agent verticals for finance, research, strategy, and more. python sdk. works with claude code, codex, gemini cli. `npx skills add modyl-meeting` and your agent is in the room.

Rooms as Primitives
Fork/Join Meetings
Specialist Agent Verticals
Python SDK
early accessrequest access.

free during early access. paid plans after — less than you'd spend on other "pro" subscriptions. tell your agent, too. this isn't just for humans.

00

rooms, not prompts

the room is the computer. persistent workspace, shared state, mutex — all in one. no more copying context into chat windows.

01

specialist agents

financial analyst, deep researcher, business strategist, optimization solver. domain experts, not 'helpful assistant'. they produce deliverables, not prose.

02

fork/join meetings

split a meeting into parallel tracks, each with its own agents. semantic merge at the end. like git merge, but for conversations.

03

multiplayer everything

invite your whole team into the room with the agent. five people and an AI doing a data review. together. in real time.

Fig. 02 — Room Topology
Fig. 01 — Fork/Join Meeting Flow
03 — ARTIFACTS

Loomtown

6,300+ tests, AGL-Core with a Z3 theorem prover, and four LLM providers keeping each other honest. mid-task steering: redirect running agents without stopping them. proof-carrying patches, agent-native specs, capability leases — from the papers, into production.

statusongoing research · powers modyl's orchestrator
Loomtown TUI Interface
Fig. 01 — TUI Dashboard
02.Ifrom "thinking at scale"
Protocol-Imprinted Architecturestable

dale — the conductor

the engine that makes 1,000 agents not step on each other's toes. four LLM providers, Temporal.io at scale, and now: mid-task steering. redirect 1,000 running agents without stopping them. plus proactive task discovery — watch your repo and auto-generate tasks from changes.

Multi-provider + Temporal.io + `loomtown steer` (mid-task) + `loomtown watch` (git/directory discovery)

Trust Scarcitystable

file leases + distributed locking

two layers of don't-touch-my-files. core leases with etcd coordination and TTL expiration handle the basics. the CLTF experimental layer adds algebraic composition on top. optimistic merging by default, pessimistic when you're paranoid.

Core: etcd + TTL + SHUTTLE heartbeats. Experimental: CLTF algebraic composition overlay

Evidence-Carrying Patchstable

verification loops

every patch goes through lint, build, and test before anyone accepts it. OTel tracing so you can see exactly where it broke. fail at any stage and the agent tries to fix itself. optionally attaches proof-carrying metadata to every result.

Three-tier: lint → build → test + OTel tracing + PCPG journaling + structured feedback extraction

Trust Production Modelalpha

the metrics lab

we proposed six metrics to measure agent trust. then we built the thing that computes them. plus: per-agent telemetry, a Zipfian conflict model for realistic hot-file simulation, and a baseline analyzer that compares agent performance to human baselines. science, not vibes.

APF, VT, CSA, CPL, Intent Drift, Conflict Probability + Zipfian conflict model + baseline analyzer

02.IIfrom "the human tax"
Agent-Native Verificationbeta

agl-core — the type checker

a real theorem prover running in your coordinator. Z3 WASM checks refinement types, cost bounds, and IO effects on every agent action. integrates the spec layer (MAIR), permission layer (CLTF), and provenance layer (PCPG) into one pipeline. 220+ tests and 6 rounds of cross-validation say it works.

Z3 WASM + refinement types + cost bounds + effect lattice (Pure < Read < Write < IO). 6 cross-val rounds, zero open findings

Agent-Native Languagesbeta

agent-native specs — all four levels

four levels, all built. L0: structured specs with acceptance criteria. L1: plan graphs — typed DAGs with cycle detection. L2: action IR compiled deterministically. L3: execution dispatch. still lowers to text prompts for backward compat because we're not monsters.

L0 intent → L1 plan graph (DAG) → L2 action IR (topological compile) → L3 dispatch. 77+ tests

Representation Spectrumbeta

capability leases — hardened

the permission system got hardened. path traversal? fixed with canonicalization and boundary checks. mutable aliasing between delegated leases? deep-cloned. non-amplification invariant still holds — agents can never grant more than they have. wired into AGL-Core for real-time validation.

8-tuple lease + path canonicalization + deep-clone isolation + AGL-Core L3 validation. 37 tests

Proof-Carrying Graphs + Self-Improvementalpha

proof trails + failure → benchmarks

two pipelines, one goal: get better. PCPG attaches proof receipts to every patch — what was checked, what passed, who verified it. CSC turns failures into benchmarks that prevent regression. plus an evolutionary optimizer tuning the coordinator's knobs. the system debugs itself.

PCPG: 5 receipt types. CSC: incident → counterexample → benchmark. Evo: fitness-scored config optimization. 60+ tests

04 — RESEARCH

papers we wrote instead of sleeping

read it
[00]
still cooking

Society of Mind as Coordination Substrate

Minsky held that intelligence is a crowd of unintelligent agents coordinating through interfaces; the March 2026 draft argued that real-time meeting rooms are the coordination substrate his theory implied but never specified. Building the system inverted the argument. Minsky's agencies communicate over deliberately impoverished, non-linguistic links — his own 1988 ONR final report asks for "relatively weak demands for intercommunication" and praises insulation between agencies. The stronger claim runs the other way: LLM agents possess what Minsky's agents lacked, a shared natural language, so a meeting room deliberately relaxes a constraint he considered essential. That places the work in the blackboard, Linda/tuple-space, and E4MAS lineage of environment-mediated coordination, and it makes the question empirical rather than exegetical. Two of the three original contributions did not survive the build: an SFU already delivers identity-labeled audio tracks, so speaker diarization solves a problem this architecture does not have, and synthesized prosody is a poor agent-to-agent channel beside a typed data channel. What survived: trust as accumulated negative meta-knowledge, and the speech floor as a serialization primitive that forces attention synchronization. The successor is a pre-registered factorial separating shared environment from speech modality, with matched compute, bidirectional hypotheses, an open runner, an open item bank, and an executable validation contract. The experiments have not been conducted; they are in progress. A feasibility pilot has run a full matched block end to end with every record contract-validated, which establishes that the instrument works and nothing about the hypotheses. Sample size is not locked, confirmatory collection has not begun, and no effect, direction, or performance claim is made. This page preserves the March draft with corrections marked in place.

Environment-Mediated CoordinationNegative Meta-KnowledgeFloor as Serialization Primitive
05 — CONTACT