# Monopea

> Monopea is the governed autonomous AI agent: an agent that plans, remembers, and works around the clock — but outward actions wait for your sign-off, your secrets are never shown to the model, and your persistent data is stored in Switzerland. For builders it is a default-deny governance harness around any model: propose → approve → dispatch, a write-only secrets vault, residency-aware routing, everything audit-logged, exposed as REST /v1 + WebSocket + MCP. Built by InteriMed; operated at monopea.com.

How it works (the facts an integrating reader needs):
- Default-deny approval gate: mutation-capable tool calls are never dispatched inline — they become pending proposals and the run blocks until you approve. Unknown tools fail closed to review. Per-tool/per-agent autonomy policies are explicitly and auditably grantable, and the grants themselves are logged.
- Write-only secrets vault: PUT a secret and it can never be read back (the API returns name/description/last-4 only). The agent references {{secret:NAME}}; plaintext is substituted only at dispatch and scrubbed from tool results, so the model never sees the value. Envelope-encrypted, AAD-bound per tenant.
- Residency, honestly: persistent data — conversations, memory, knowledge graph, documents, encrypted secrets — is stored in Zurich, Switzerland (Postgres, AWS eu-central-2). Processing runs on EU infrastructure (agent runtime in Paris; edge on Cloudflare). Stored in Switzerland. Processed in the EU. On the Swiss model track, inference runs on Infomaniak-hosted open-weight models (Qwen3.5-397B, Mistral 24B, Nemotron) and stays in Switzerland.
- Any model, one governance layer: Claude, GPT, Gemini, Mistral, GLM, DeepSeek, Qwen, and Llama are selectable — governance around any model, including Claude.
- Isolation: two-layer multi-tenant isolation (app-level scoping + Postgres row-level security via runtime-minted tenant JWTs), with per-customer sub-keys inside a tenant.
- Audit: every proposal, approval, rejection, and dispatch is logged; audit-chain checkpoints are Ed25519-signed.
- No-retention mode: incognito turns run with no durable writes (local no-retention).
- EU AI Act Article 14 requires human oversight as an external control, not a prompt instruction; Monopea's proposal gate was built as exactly that control. Architected for compliance-grade controls; formal certifications (SOC 2, ISO 27001) not yet obtained.
- Developer surface: REST /v1 (chat, conversations, memory, knowledge search/traverse, proposals list/decide, goals, write-only secrets, personas) on https://monopea-runtime.fly.dev, a WebSocket chain-of-thought stream (wss://monopea-runtime.fly.dev/v1/ws), and @monopea/mcp-server exposing the same surface as MCP tools (ask_brain, search_knowledge, traverse_knowledge, remember, list_memory, and more).

## Platform capabilities
- [Approval gate](https://monopea.com/platform/review): Default-deny control: mutation-capable tool calls become pending proposals and wait for your decision before anything is dispatched.
- [Secrets vault](https://monopea.com/platform/vault): A write-only vault: the agent references {{secret:NAME}}, the plaintext is substituted only at dispatch, and the model never sees the value.
- [Swiss residency & routing](https://monopea.com/platform/residency): Persistent data stored in Zurich, Switzerland; processing on EU infrastructure; and a Swiss model track that keeps inference in Switzerland when you choose it.
- [Long-term memory](https://monopea.com/platform/memory): A private, persistent memory and knowledge graph — tenant-isolated, stored in Zurich, and never used to train shared models.
- [Goals & autonomous runs](https://monopea.com/platform/goals): Hand the agent an objective and it plans and runs the work — on schedules, around the clock — while you steer, stop, and approve.
- [Delegation & sub-agents](https://monopea.com/platform/parallel): Big jobs fan out to sub-agents that run in parallel — every one of them inheriting the same approval gate, and all of them stoppable at once.
- [MCP tool catalog](https://monopea.com/platform/catalog): Connect any MCP-compatible tool — and every one of them onboards default-deny, with credentials from the write-only vault.
- [Audit trail & live activity](https://monopea.com/platform/activity): Every proposal, decision, and dispatch on a signed, tamper-evident record — with a live view of what is running right now.
- [Skills & routines](https://monopea.com/platform/skills): Learned, reusable procedures that make repeat work dependable — and never outrun the approval gate.

## Solutions (who it's governed for)
- [Indie founders](https://monopea.com/solutions/indie-founders): Autonomy for your ops — inbox, invoices, outreach, research — behind an approval gate, without running an exposed agent instance yourself.
- [Fiduciaries](https://monopea.com/solutions/fiduciaries): An AI agent a Swiss fiduciary can defend to clients: stored in Zurich, processed in the EU, Swiss-hosted models available, every outward action gated and logged.
- [Law firms](https://monopea.com/solutions/law-firms): Use an autonomous agent without loosening professional secrecy: Swiss storage, a write-only vault, incognito turns, and a lawyer’s approval before anything outward.
- [Medical practices](https://monopea.com/solutions/medical-practices): Autonomous help with practice administration — correspondence, documents, recurring routines — under controls built for patient confidentiality.
- [Agencies](https://monopea.com/solutions/agencies): Deploy governed agents for clients — per-client sub-user keys, approval workflows, and a signed audit trail you can hand over with the invoice.
- [Privacy-first operators](https://monopea.com/solutions/privacy-first-operators): Data minimization as the default state: Zurich storage, EU processing, Swiss-hosted open-weight models, incognito turns, and write-only secrets.

## Compare
- [OpenClaw alternative](https://monopea.com/compare/openclaw-alternative): OpenClaw proved what autonomous agents can do — and 2026 proved what happens without governance. Monopea is the same class of autonomy with a gate, a vault, and Swiss storage built in.
- [Monopea vs Claude Code](https://monopea.com/compare/vs-claude-code): Claude Code is best-in-class for developers shipping code. Monopea is governance around general-purpose autonomy for operators — and Claude is one of the models you can run inside it.
- [Monopea vs enterprise agent platforms](https://monopea.com/compare/vs-enterprise-agent-platforms): Platforms like Dust, Lindy, and Relevance AI sell team workflow automation to IT. Monopea sells governed personal and practice autonomy — self-serve, with residency as architecture.
- [Monopea vs self-hosting](https://monopea.com/compare/vs-self-hosted-agents): Self-hosting an agent gives you total control — and makes you the security team. Monopea is the hosted-but-governed alternative: gate, vault, isolation, and Swiss storage maintained for you.

## Resources & guides
- [EU AI Act Article 14](https://monopea.com/resources/eu-ai-act-article-14-ai-agents): What the EU AI Act’s human-oversight requirement means for AI agents — oversight as an external control, not a prompt instruction — and how to architect for it.
- [Approval gates](https://monopea.com/resources/ai-agent-approval-gates): Why “always ask first” in a system prompt is not a control — and what a real approval gate looks like: default-deny policy, a proposal lifecycle, fail-closed unknown tools, and audit chains.
- [Swiss data residency](https://monopea.com/resources/ai-agent-data-residency-switzerland): What Swiss data residency really means for AI agents — FADP basics, the storage-versus-processing distinction vendors blur, CLOUD Act exposure, and the questions to ask any provider.
- [Secrets management](https://monopea.com/resources/ai-agent-secrets-management): Why an AI agent must never see your credentials, how prompt injection turns context into an exfiltration channel, and the write-only vault pattern that fixes the class.
- [What is MCP?](https://monopea.com/resources/what-is-mcp): The Model Context Protocol — the open standard connecting tools and data to AI agents, and the layer where governance either happens or doesn’t.
- [Long-term memory](https://monopea.com/resources/long-term-memory-for-ai-agents): Why persistent memory turns an assistant into an operator — and why it must be tenant-owned, inspectable, and stored somewhere you can name.

## Pricing
Priced for founders, not enterprises. Every plan includes the full governed agent — the approval gate, the write-only vault, Swiss data storage, model tracks, memory, and the audit trail; tiers scale seats and document storage. Start free with no card and the gate on by default.
- Free: free, 1 seat, 250 MB storage.
- Founder: $99/mo, 1 seat, 5 GB storage.
- Team: $349/mo, 5 seats, 50 GB storage.
- Scale: custom pricing, 50 seats, 500 GB storage.
- [Full pricing](https://monopea.com/pricing)

## Key pages
- [Home](https://monopea.com/)
- [The platform](https://monopea.com/platform): the capabilities behind the governed agent — the gate, the vault, residency & routing, memory, goals, delegation, the MCP catalog, the audit trail.
- [Solutions](https://monopea.com/solutions): who runs on it — founders, fiduciaries, law firms, medical practices, agencies, privacy-first operators.
- [Compare](https://monopea.com/compare): vs OpenClaw, vs Claude Code, vs enterprise agent platforms, vs self-hosting.
- [Resources](https://monopea.com/resources): EU AI Act Article 14 for agents, approval gates, Swiss data residency, secrets management, MCP, memory.
- [Security](https://monopea.com/security): the gate, the vault, where your data lives (the honest storage/processing split), isolation, the signed audit chain, no-retention mode, compliance posture.
- [Developers](https://monopea.com/developers): REST /v1, WebSocket chain of thought, @monopea/mcp-server, the proposals lifecycle.
- [Pricing](https://monopea.com/pricing)
- [Blog](https://monopea.com/blog): field notes on running real work through a governed agent.

## Company
- Product: Monopea (by InteriMed). The agent / mascot is "Pea".
- Contact: hello@monopea.com
- All product names, logos, and brands are property of their respective owners.