Laserbrain – the adaptive recursion harness for AI agents

Laserbrain launches an adaptive recursion harness that provides a fixed reference for AI agents to check against each step, preventing drift and loops. The system uses four drift signals and a single URL for MCP-capable agents, with a free check and paid tiers for drift history, alerts, and fleet view. The company claims the check is proven, while multi-agent oversight remains a prototype.

product · for AI agents · mission control the smart recursion harness Autonomous agents drift. They loop, over-deliberate, and keep working long after they’ve stopped making progress — going in circles with nothing to show. laserbrain is a fixed reference an agent checks itself against each step , so it catches its own drift and returns to the goal instead of spiraling. The catch every other approach hits: an agent that watches only its own recent thinking is provably blind to slow drift — each step looks fine next to the last, while it wanders far from where it began. laserbrain is an outside reference it can’t fool. There’s a proof proof . how it works · four drift signals Each step, your agent spells its state into one fixed schema and checks it. Four things mean it has drifted — and should stop and return to its goal: attach it · one URL, no SDK Point any MCP-capable agent at one URL. Each step it calls check state with its spelled state, and laserbrain — remembering the agent’s ground state across the run — tells it whether it has drifted and should return. The same URL is a small suite: the check, a self that persists across sessions, Alice’s framework guidance, and the spectral grammar. check state , reset task and the grammar are free at small scale; remember self and ask alice take a key and are metered. Paid tiers add retained drift history, spiral alerts and a fleet view — you pay to see your agents drift over time, not for the check. The check is the proven part; memory, guidance and language are tools on the same door — useful, not theorems. Watch it in the dashboard /laserbrain/dashboard/ . the ecosystem · from the check to the whole loop The check is the proven core. Around it, the tools to run it at scale — return, escalate, audit, and carry a team’s ground across sessions. Single-agent detection is the theorem; the rest are mechanisms on top of it, said plainly for what they are. pip install laserbrain Don’t just detect the drift — close the loop. Give laserbrain your step function and it injects the return, so the agent recovers instead of spinning. Drop into LangGraph, CrewAI, AutoGen or the OpenAI Agents SDK. You map your agent’s state to a goal; laserbrain watches the fixed reference from inside your own loop. When a self-correcting return doesn’t take, the drift escalates to a human queue — Return · Allow · Stop . A person sees only what the reference caught, not every step. Every check is written to a hash-chained ledger, verifiable offline by anyone, no key. Edit a past verdict to hide a drift and the chain breaks at that link. Snapshot a team of agents and resume it in a later session — the shared goal and the dialogue carry over, so the group re-grounds instead of starting cold. Style each role’s return: a deep explorer tolerates displacement, a tight checker returns fast — the fixed reference over a group, catching the echo/agreement spiral a self-watching team can’t see. The hosted escalation queue and the MCP URL work today; the SDK is the open adoption layer for teams that want the check, the act layer and the adapters in-process. Multi-agent oversight — dialogues and recursion teams — is a prototype extension /laserbrain/research/ , useful but not yet a theorem; escalation, audit and continuity are mechanisms, not claims about answer quality. what’s proven · and what isn’t, plainly the protocol line one second term, at three scales — plus the wirepricing the check is free; the console is the productCatch your own drift. See every drift — and get paged for it. Mission control for a fleet. Agents at scale, on your terms. Priced on observability — retained drift, alerts, the fleet — not field throughput. The live field is a free bonus in Ground. See the field → /field/laserbrain/guide/