How Zino works.
The technical reference for the platform and security engineers who will run Zino: where it sits, what it intercepts, how credentials move, what it plugs into, and what a four-week pilot looks like.
At a glance
- Where it runs
- In-process SDK in each workload. No sidecar, no network hop for policy.
- Identity
- A workload identity from SPIRE, checked by two-layer attestation.
- Policy
- Signed WASM policy from one Policy Server, evaluated on the OPA schema.
- Intercepts
- REST/HTTP, Shell and MCP over stdio, SSE and HTTP.
- Secrets
- Brokered from your vault and injected per call. The agent holds none.
- Audit
- Every decision is written to a hash-chained log.
Two planes: one decides, one enforces.
The Policy Server is the single signing authority for the fleet. The Zino SDK runs in-process at every workload, with no sidecar and no network hop for policy evaluation, gating REST, Shell, and MCP calls at the trust boundary before anything leaves.
Three surfaces. One policy interface.
REST, Shell, and MCP all speak different dialects. Zino intercepts all three with a single policy engine.
- Surface
- REST / HTTPrequests · httpx · aiohttp
- What it sees
- A CallContext (method, host, path, tool, and arguments) built for every outbound HTTP call.
- Policy decides
- Re-authorized on the same OPA schema. A call to an unapproved host is denied before it is sent.
- Surface
- Shelltool-type aware
- What it sees
- Execution type classified as HTTP, shell, or MCP; shell commands registered straight from policy.
- Policy decides
- Pre-execution callbacks can inspect or veto before any command runs.
- Surface
- MCPclient-session wrap
- What it sees
- Tool name, schema, and arguments on every outgoing MCP call, across stdio, SSE, and HTTP transports.
- Policy decides
- Each call authorized before it leaves the agent, closing the gap to un-vetted MCP servers.
Zino controls secrets, not the app.
Tokens are injected at call time, configuration is encrypted at rest, and your enterprise vault stays the source of truth.
Credential broker · platform-controlled, per call
App makes an outbound call. Any Authorization header it set is stripped.
The interceptor fetches the credential scoped to (host, path, method, tool).
The platform token is injected, decrypted in memory via a per-agent HKDF key.
The agent never holds a long-lived API token.
Fronts your enterprise vaults
- HashiCorp Vault
- OpenBao
- AWS Secrets Manager
- Azure Key Vault
- Google Cloud
- CyberArk Conjur
- Thales
- Infisical
Framework adapters
Zino ships adapters for these agent frameworks. One SDK call instruments an agent for policy enforcement and audit.
- LangChain
- CrewAI
- AutoGen
- Claude Tool Use
- LlamaIndex
- OpenAI Assistants
- Semantic Kernel
Four weeks. One agent fleet. Real policy, enforcing.
A bounded pilot with one sponsor and no infrastructure lift. Walk away with identity, attestation, and a real authorization layer in production.
Week 0
Setup
Policy Server in your environment. Trust root provisioned. SPIRE Agent rolled to one fleet. No app changes yet.
Week 1
Identity
One initialize() in the pilot agent. Two-layer attestation working end-to-end. Shadow mode: observe only.
Week 2
Policy
Interception live across REST, Shell, MCP. Allow/deny enforced. Audit log streaming. Slack alerts on blocks.
Week 3
Callback
Pre-execution callbacks wired in. App-specific rules enforced. Joint readout + rollout plan for the next two teams.
Pilot commitment · 4 weeks · 1 sponsor · 1 fleet · no infrastructure lift
