Skip to content
API Blog

Comparison

Lobu is the open-source backend for AI teammates that watch, remember, and act. Connectors and webhooks build a live org knowledge graph; agents look it up and branch into a sandbox to do work — with per-channel isolation, OAuth, and credentials agents never see.

Prefer a shorter product-facing overview? See Compare lobu (OpenClaw, Claude Tag, Viktor, Town).

This page is the deeper technical comparison against other ways to run agents for multiple users, including Claude Tag (@Claude in Slack).

LobuOpenClaw (direct)Claude Tag
What it isOpen-source backend for multi-user agentsSingle-tenant agent runtime@Claude in Slack (team agents, beta)
Multi-tenantPer-user/channel isolationSingle-tenant — shared filesystemPer-channel @Claude
PlatformsSlack, Telegram, WhatsApp, Discord, Teams, Google Chat, REST API, MCPCLI and APISlack (beta)
EmbeddableMount inside Next.js, Express, Hono, FastifyNoNo
Self-hostedSingle Node process (embedded Postgres, or BYO Postgres)Single processCloud only
Model supportAny provider via configAny providerAnthropic only
RuntimeOpenClawOpenClawClaude
Network isolationGateway-mediated egress, domain filteringHost networkPlatform-managed
Secrets handlingGateway proxy, workers never see credentialsDirect env varsPlatform-managed
MCP supportProxied through gateway with secret injectionDirectYes
Agent Protocol / A2ANot yetNoNo
Built-in evalsYAML eval framework with model comparisonNoNo
MemoryAppend-only org graph (connectors, entities, automations)Local filesystemChannel-scoped, admin-provisioned
Custom connectors / automationsYes (lobu.config.ts)Skills + local setupAdmin-provisioned tools only
Worker lifecyclePersistent subprocess per channel; reaped on config change / shutdownAlways runningManaged
Config formatlobu.config.ts (TypeScript) + IDENTITY/SOUL/USER.mdCLI flagsDashboard
LicenseOpen sourceOpen sourceProprietary

Lobu’s bundled memory system is benchmarked against Mem0 and Supermemory on public datasets. Same answerer (glm-5.1 via z.ai), same top-K, same questions.

Single-session knowledge retention.

SystemOverallAnswerRetrievalLatency
Lobu87.1%78.0%100.0%237ms
Supermemory69.1%56.0%96.6%702ms
Mem065.7%54.0%85.3%753ms

Multi-session conversational memory.

SystemOverallAnswerRetrievalLatency
Lobu57.8%38.0%79.5%121ms
Mem041.5%28.0%66.9%606ms
Supermemory23.2%14.0%36.5%532ms

See the memory benchmarks methodology for fairness guardrails and reproduction steps.

Lobu runs as a single Node process. Each user/channel gets its own worker subprocess that the gateway spawns on demand.

Uses just-bash (virtual bash) + Nix for reproducible packages. Each user gets an isolated virtual filesystem and bash session. Resource use and concurrency depend on workload, model calls, and host capacity, so benchmark the deployment shape you plan to run.

  • Subprocess boundary per userchild_process.spawn per session; SIGKILL recoverable.
  • Workspace persistence./workspaces/{agentId}/ survives gateway restarts.
  • Egress through gateway proxy — workers run with HTTP_PROXY=http://localhost:8118; allowlist/blocklist + LLM egress judge enforced at the proxy.
  • Linux production hardening — when systemd-run is available, the worker spawn becomes a transient unit with MemoryMax, CPUQuota, IPAddressDeny=any (kernel-level egress), capability drops, and NoNewPrivileges. macOS dev hosts fall back to plain spawn.
LobuE2B
IsolationSubprocess + just-bash + isolated-vm + (Linux) systemd-runFirecracker microVM
Self-hostedYes (single Node process)Cloud API
Network controlGateway proxy + domain filtering + IPAddressDeny on LinuxSandbox-level
Startup timeInstant (in-process gateway, subprocess workers)<200ms
PersistenceDirectory per agent24hr max
Per-user isolationYes (per-channel/DM subprocess)Per-sandbox

Workers call MCP tools through the gateway. The gateway resolves ${env:VAR} secrets and injects OAuth tokens before forwarding to the upstream MCP server. Workers never see credentials — they receive opaque proxy URLs.

LobuClaude TagDirect MCP
Secret injectionGateway proxy resolves at request timePlatform-managedDirect env vars
OAuth managementLobu handles token refreshPlatform-managedManual
Transport supportHTTP, SSE, stdio (proxied)YesAll
Worker sees credentialsNeverN/AYes
Audit trailGateway logs all MCP callsNoNo

For compliance-bound deployments, agent code never touches API keys or OAuth tokens — even if the agent is compromised.

Hosted platforms (Claude Tag and similar) require sending your data, prompts, and agent memory to a third party. For regulated industries (finance, healthcare, government) or organizations with data residency requirements, this is a non-starter.

Lobu runs entirely on your infrastructure:

  • Data stays in your network — Postgres, workspaces, and memory are all self-hosted
  • No external dependencies — the gateway, workers, and MCP proxy run on your machines
  • Network-level isolation — workers use gateway-mediated egress, with kernel-level loopback-only enforcement on Linux production hosts
  • Credential separation — secrets never leave the gateway process
  • Audit everything — gateway logs all LLM calls, MCP tool invocations, and network requests
  • Air-gapped compatible — with local LLM providers (Ollama, vLLM), Lobu can run fully disconnected

Use Lobu when your agent needs more than one isolated user or channel:

  • SaaS products — embed agents in your app where each user gets isolated persistence, tools, and context.
  • Internal teams — deploy a single bot to Slack or Teams where every employee gets their own sandboxed agent.
  • Customer support — agents that handle tickets autonomously with human-in-the-loop approval gates.
  • Managed agent services — operate agents for clients with per-tenant isolation and network controls.

If you need a single personal agent for yourself, use OpenClaw directly.

Lobu and OpenClaw are complementary. OpenClaw is the single-user runtime. Lobu adds multiplayer infrastructure around it: routing, isolation, credentials, memory, and delivery.

OpenClaw (~800k LOC) was designed as a single-tenant, single-user system. Production deployments need multi-tenant isolation, platform routing, credential separation, and network control — concerns OpenClaw doesn’t have opinions about.

CapabilityLobuOpenClaw
ArchitectureMulti-tenant gateway + isolated workersSingle-tenant, single-user
Platform deliverySlack, Telegram, WhatsApp, Discord, Teams, Google Chat, REST APICLI and API
Worker isolationSubprocess + just-bash + (Linux) systemd-run hardeningRuns on host
Secret handlingGateway proxy injects credentialsDirect env vars
Egress controlDomain allowlists via HTTP proxyHost network
Worker lifecyclePersistent subprocess per channelAlways running
DeploymentSingle Node process (embedded Postgres or BYO Postgres)Single process

Inside each Lobu worker, the full OpenClaw runtime runs untouched. Lobu rewrites only the gateway layer (~40k LOC) to be multi-tenant.

What “we’ll build it ourselves” entails:

  • Sandboxing: per-user worker lifecycle with workspace persistence
  • Platform adapters: Slack Events API, Telegram, WhatsApp webhooks, each with their own auth flows
  • Credential isolation: proxy layer that injects secrets without exposing them to agent code
  • Network policy: domain-filtered egress through a gateway proxy
  • MCP proxy: secret injection, OAuth token refresh, and routing for MCP servers
  • Eval framework: automated quality testing across models
  • Admin UI: per-agent configuration, connection management, status monitoring

Lobu handles all of this out of the box. The gateway is ~40k lines of TypeScript that took months to build and harden.