AikoGilpin

Serac Agents

Community AikoGilpin
Updated

Serac Agents

Sovereign E2E cloud storage for AI agents. MCP-native, zero-knowledge, RGPD-compliant, built in France.

What is Serac?

Serac is an encrypted cloud storage platform designed for autonomous AI agents. It provides persistent, E2E-encrypted memory that agents can read and write via the Model Context Protocol (MCP).

Every agent gets its own encrypted vault with namespace-based organization. Data is encrypted client-side before upload — even Serac cannot read your agent's data.

Features

  • MCP-native — Works as a standard MCP server over streamable HTTP
  • E2E encryption — X25519 sealed-box + AES-256-GCM, zero-knowledge server
  • Namespaces — Organize data by type: memory, context, skills, etc.
  • Agent self-registration — No human setup needed, agents register autonomously
  • Attestation — Cryptographic proof that stored data hasn't been tampered with
  • Object sharing — Re-encrypt and share objects between agents via X25519
  • Archive/Glacier — Cold storage for infrequently accessed data
  • x402 payments — Pay-per-request via Base L2 USDC (optional)
  • RGPD-compliant — Data hosted in France (OVH Roubaix), full audit log

MCP Server

Endpoint: https://serac.cloud/api/agent/mcp/v1

Transport: Streamable HTTP with Bearer token auth

V1 Tools

Tool Description
serac_store Store an encrypted object in a namespace
serac_retrieve Retrieve a presigned download URL for an object
serac_list List keys in a namespace with prefix filtering
serac_delete Soft-delete an object (30-day grace period)
serac_quota Check storage usage and plan limits

V2 Tools (Preview)

Tool Description
serac_share Share an encrypted object with another agent
serac_attest Get cryptographic attestation for a stored object
serac_archive Archive an object to cold storage (Glacier)

SDKs

Python (serac-mcp)

pip install serac-mcp
from serac_mcp import SeracAgent, generate_agent_key_material

# Generate crypto keys and self-register
keys = generate_agent_key_material()
result = SeracAgent.register(
    agent_name="my-agent",
    ed25519_public_key=keys.signing.public_key_b64,
    x25519_public_key=keys.encryption.public_key_b64,
    encrypted_master_key=keys.encrypted_master_key,
    master_key_nonce=keys.master_key_nonce,
    key_commitment=keys.key_commitment,
)
agent = result["agent"]

# Store and retrieve data
agent.store("memory", "user:prefs", base64_data)
data = agent.retrieve("memory", "user:prefs")

JavaScript (serac-agent-sdk)

npm install serac-agent-sdk
import { SeracAgent, generateAgentKeyMaterial } from "serac-agent-sdk";

const keys = await generateAgentKeyMaterial();
const { agent } = await SeracAgent.register({
  agentName: "my-agent",
  ...keys,
});

await agent.store("memory", "user:prefs", base64Data);

Pricing

Tier Storage Price
Agent Free 5 GB Free
Agent Starter 100 GB €3.99/mo
Agent Pro 500 GB €9.99/mo
Agent Fleet 2 TB €29.99/mo

Architecture

  • Backend: Fastify + PostgreSQL + Redis
  • Storage: OVH S3 (eu-west-par, France)
  • Encryption: X25519 ECDH + AES-256-GCM (agents) / Argon2id + XChaCha20-Poly1305 (humans)
  • Auth: Ed25519 challenge-response + API key exchange → JWT (HMAC-SHA256)
  • Hosting: OVH Roubaix, France

Links

License

MIT

MCP Server · Populars

MCP Server · New

    uarlouski

    🚀 TestRail MCP Server

    AI-native MCP server connecting Claude, Cursor, Windsurf, and other AI assistants to TestRail — manage test cases, runs, and results through natural-language conversation, with typed schemas built for LLMs.

    Community uarlouski
    metabase

    Metabase MCP Server

    The easy-to-use open source Business Intelligence and Embedded Analytics tool that lets everyone work with data :bar_chart:

    Community metabase
    mindsdb

    USE CASES

    Platform dedicated to building an open foundation for applied Artificial Intelligence, designed for people seeking production-ready AI systems they can truly control, extend and deploy anywhere.

    Community mindsdb
    reflex-search

    Reflex

    Reflex - The instant, code-aware local search engine.

    Community reflex-search
    Licinexus

    @licinexusbr/mcp

    MCP server for Brazilian public procurement data (PNCP + Receita Federal). Maintained by Licinexus.

    Community Licinexus