Skip to main content
WalletSuite builds, signs, and operates the hosted MCP runtime. There is no package to install and no process to run on your side — your agent connects to the hosted endpoint. The build pipeline, the attestation, and the running process are WalletSuite’s responsibility. This page covers the supply-chain and runtime controls behind that surface.

​
Attested Hosted Runtime

The runtime that answers at the hosted MCP endpoint is built from source WalletSuite controls, on a pipeline WalletSuite operates. Each build is attested to the exact source commit and workflow run that produced it, and only attested builds are promoted to the hosted runtime. You consume the endpoint, not a tarball — nothing to verify, nothing to keep patched on your side.

​
Runtime Response Validation

Every backend response consumed by a tool is validated against a typed contract before it reaches the model. A tampered or unexpected response fails schema validation and surfaces as a structured, sanitized error — it never becomes input to agent reasoning. The same validation boundary applied to tool arguments is applied symmetrically to the other side of the API call. Token amounts are preserved losslessly as strings. Large integer values are never coerced to floats, eliminating a class of precision-loss issues on balances and transfer amounts.

​
Prompt Injection and Tool Hygiene

The hosted runtime treats the model as an untrusted caller. What a model can even attempt is bounded first by the tool profile: tools outside the agent’s granted profile are never registered in the MCP schema, so the model cannot discover or call them no matter what a prompt says. Inside that surface:
  • Every tool input is validated against a Zod schema before any backend call. Address format, amount format, and mutually exclusive fields are enforced at the boundary.
  • Every backend response is validated against a handwritten contract. A tampered or unexpected response fails schema validation and becomes a structured error, not agent input.
  • Internal diagnostics are logged server-side and sanitized before reaching the model. The model sees a structured, user-safe error; the runtime retains the detail.
  • Logs use operation labels rather than URLs. Wallet addresses, transaction parameters, and response payloads are not written.

​
Secret Boundary

On the hosted path the customer’s only secret is the agent’s key. It stays on your server and authorizes the agent; the runtime only ever sees a short-lived token, never the key, and the key never reaches the model, the browser, or a prompt. See Connect Your Agent. Signing is non-custodial MPC threshold signing: no full private key is ever assembled, so there is no complete signing key for the hosted runtime to load, store, or expose. WalletSuite’s share is envelope-encrypted at rest behind the scenes; External BYO signing (customer-managed KMS / HSM / multisig) stays available. See the trust model. See Security Overview → Trust Model for the custody model and full credential lifecycle.

​
Logging and Redaction

Logs are designed to be useful for forensics without leaking request-identifying material.
  • Operation labels replace URLs. Wallet addresses, transaction payloads, and request parameters are not logged.
  • Internal diagnostics are retained server-side. The model receives a structured, sanitized error; the operator’s logs retain the detail.
  • Audit receipts use schema-driven redaction: only dynamic text fields are scanned for inline secret patterns. Operational fields — wallet name, chain, identifiers, hashes, policy IDs — are preserved for audit, compliance, and forensic review.
  • Transaction hex is truncated before write, so even a serialized payload containing key material would not land in storage intact.
See Audit Trail for the receipt schema and hash chain.

​
Dependency Surface

WalletSuite operates the build pipeline for the hosted runtime. A release that carries a high or critical advisory against a runtime dependency does not ship — the build is blocked before it reaches the hosted endpoint. For the broader posture on formal audits and third-party certifications that sit above the build pipeline, see Security Diligence → Certifications Roadmap.