LabMission

The lab platform for sample management, LIMS and lab operations

Applications your lab runs from day one, on an enterprise platform that your team, your AI agents, or both can configure, connect and extend, with every change governed and on the record.

  • Sample Management
  • LIMS
  • Lab Operations
  • Biobanking
  • MCP server
  • Workbenches
LabMission · Sample Management · Receive
LabMission Sample Management: incoming shipments triaged into ready, needs attention, backlog and problem, with notifications open.
One product

Applications on top. A governed platform underneath.

The applications are where your lab starts. The platform under them is what lets LabMission grow into whatever your lab needs next: new instruments, new studies, new sites, new tools.

Built and run by
  • People
    The canvas and the applications
  • AI agents
    The MCP server
  • Code
    The API, SDK and workbenches

All three work through the same proposals, rights and record.

The platform under every application
Connect
  • Connections
  • API sources
  • Simulators
  • Files & manifests
Data
  • Live streams
  • Queries
  • Data stores
  • History & export
Build
  • Surfaces
  • Interfaces
  • Agents
  • Workbenches
Govern
  • Provenance
  • Access rights
  • E-signatures
  • Change control
REST API & OpenAPI specMCP serverTypeScript SDKAPI keys
Your cloud
A workspace per tenantSecrets in your key vaultImmutable retentionData-residency policy
A starting point

Start with sample management. Keep building.

Sample management is the first application, not the edge of the product. The same workspace that runs it is where your lab adds what it needs next, without a separate project or a separate system.

  • 1Connect a new plate reader, freezer or robot as configuration
  • 2Add a study type, a site or a workflow as a setup, not a release
  • 3Build a dashboard over any query, and publish the query as an API
  • 4Package your own code as a workbench and run it in your workspace
  • 5Bring the LIMS, ELN and data stores you already run into one record
A scientist at a lab bench working on a laptop beside her instruments.
Built by people, AI, or both

Your team or an AI agent can design, build and maintain it

Everything in LabMission can be done by hand. An AI agent can do the same work through the MCP server, and its changes arrive as proposals. AI is optional, and it is switched off until you turn it on.

Your team builds it

Configure studies, connections, queries and dashboards on the canvas and in the applications. No AI required.

An agent builds it

An AI agent reads the workspace through the MCP server and drafts the integration, query or surface as a proposal.

Both, together

Ask for a change in plain language, adjust what the agent drafted, then approve it like any other change.

  1. 1Draft
  2. 2Review
  3. 3Approve & sign
  4. 4Deploy
  5. 5On the record

One path to production, whoever drafted the change.

Agents & MCP

What an AI agent does on LabMission

An agent signs in through the MCP server with the rights you give it. It reads the workspace, queries and streams the data, and drafts new integrations, queries and dashboards. A person approves each one before it changes anything.

  • Works with Claude, ChatGPT, Microsoft Copilot or your own agent
  • Reads only what its rights allow, with secrets stripped
  • Writes only proposals, which a person reviews and signs
  • Every call and every change is on the record
See what agents can do
Your agent → LabMission MCP server
  • 01get_workspace_flow→14 nodes · 2 surfaces · 0 pending
  • 02list_data_connections→Freezer B, PR-2, Courier tracking
  • 03run_warm_queryfreezer_temps_24h→1,440 rows
  • 04tail_telemetryfreezer-b · 60 s→60 readings, live
  • 05manage_proposalscreate→#128 awaiting review
  • Signed in as agent · rights: Biobank operator (read) + propose
Assistant · Biobank · Site A
Connect the new plate reader and put its results against the samples they came from.
I read the workspace and the reader's last export. I've drafted three changes for your review. Nothing is live yet.
Proposal #128Awaiting review
  • + connection Plate reader PR-2 (HTTP)
  • + mapping    PR-2 columns → sample, well, OD
  • + query      results_by_sample
  • ~ surface    Assay results (adds 1 panel)
Reason: new instrument, change request CR-2026-114 · Two-person approval required
Approve & signRequest changes
Connections

What connects to LabMission, and where the data goes

Instruments, freezers, sensors, robots, couriers and the systems you already run feed one governed data stream. Applications, dashboards, agents and your own systems read from it.

YOUR LABCONNECTSTREAMSERVEUSEInstruments & readersResult files · HTTPFreezers & sensorsMQTT · IoT hubLiquid handlers & robotsHTTP · WebSocketLIMS, ELN & data storesEvents over HTTPCouriers & partner APIsPolled on a scheduleManifests & spreadsheetsFiles mapped on arrivalConnectionMQTT · HTTP · WebSocketIoT hubAPI sourceScheduled pollingBearer · API key · BasicSimulatorTest a flow beforethe hardware arrivesOne data streamEvery readingtime-stamped,attributed to itssource, andon the recordHotLive to screensand agentsWarmQueries, publishedas APIsColdArchive, CSV andJSONL exportApplicationsDashboardsAI agentsYour systemsEach connection is configuration, added on the canvas, by an agent's proposal, or from code, and recorded like any other change.
See connections & data
Enterprise platform

The controls an enterprise lab platform needs

These belong to the platform, so every application, every agent and every workbench inherits them.

Workspaces and teams

A workspace per tenant, with teams and members you manage.

Access rights

Rights, configurations of rights, and the people who hold them, down to which tables someone can read.

Your identity provider

Sign-in through your own directory. No passwords held here.

Provenance

Every event, amendment and signature hashed into a chain you can verify.

Electronic signatures

Six signature meanings, re-authentication before signing, and two-person approval.

Change control

Drafts, proposals, commits and deploy runs, with the history of each.

Immutable retention

Records kept in locked, extend-only storage, for periods up to 400 years.

Your cloud

Evaluation, fully managed, or in your own cloud subscription with a landing zone.

APIs and SDK

A REST API with a generated OpenAPI spec, a TypeScript SDK and API keys.

Health and telemetry

Health and inventory for every resource, and telemetry your tools can read.

Archive and restore

Data archives and restore, and export of anything, at any time.

Governed AI

The same rights, proposals and record for agents as for people.

A scientist pipetting into tubes in a rack.
An automated liquid handler with rows of pipette tips.
A scientist in a lab coat working at a cabinet.

See LabMission running

We'll walk your sample lifecycle in the application, then show the platform underneath: the connections, the queries, an agent's proposal, and the record it leaves.