Vouch Protocol MCP server

Vouch Protocol™

Vouch Protocol™ is a common-law trademark.

---

Quick Start

# Install (Linux and macOS). On Windows, use: pip install vouch-protocol
curl -fsSL https://vouch-protocol.com/install.sh | sh

# Run vouch with no arguments and pick what you want to do
vouch

Prefer to do it by hand? It is three commands:

pip install vouch-protocol

# One command to configure SSH signing + Vouch branding
vouch git init

# All future commits are now signed and show ✅ Verified on GitHub
git commit -m "Secure commit"

Setting up an agent instead of git signing? vouch onboard --quick generates a working identity, allow-list, verifier, and heartbeat config in one command.

---

What's New in v1.0

Vouch Protocol v1.0 aligns directly with the open standard:

• Verifiable Credentials as the credential format (replacing v0.x JWS tokens).
• Data Integrity proofs with the eddsa-jcs-2022 cryptosuite (no JOSE, no Base64-wrapped payload, the credential remains human-readable JSON).
• Multikey verification methods in DID Documents (algorithm-agnostic, ML-DSA-44 ready).
• Post-quantum profile as a Data Integrity proof set: the credential carries two independent proofs, eddsa-jcs-2022 and mldsa44-jcs-2024, an optional add-on for regulated deployments aligning with NIST CNSA 2.0 / NSM-10 timelines.
• Three-way cross-implementation interop verified across Python, TypeScript, and Go.

Credentials are issued with Signer.sign() and checked with Verifier.verify(). See the Specification at vouch-protocol.com/specs/SPEC/ for the full specification.

---

The Open Standard for AI Agent Identity & Accountability

When Anthropic launched MCP, they solved "how agents call tools." They didn't solve "how we TRUST those agents."

Vouch Protocol is the SSL certificate for AI agents.

Vouch is not one tool, it is a set of them. Here is the whole map.

On the command line

• vouch init generate an agent identity (a DID and keypair).
• vouch sign / vouch verify sign a payload and verify it.
• vouch git sign every git commit cryptographically, set up in one command, with a verified badge for your README.
• vouch scan find leaked Vouch key material in your code before it ships (a private key in a file, a seed in an env var, a DID document that accidentally carries a private key).
• vouch attribute separate who wrote which line. When an AI assistant and a human both edit a file, this records the AI's lines under the AI's own key and your lines under yours, so when a line causes an incident you can prove which of you wrote it. See the Claude Code integration.
• vouch media sign images, with C2PA support.

For your agents

• MCP server (vouch-mcp) a standalone Model Context Protocol server so any MCP client (Claude Desktop, Cursor, any agent) can create an identity, sign and verify credentials, scan for leaked keys, and decode DIDs, out of the box.
• Identity Sidecar keeps signing keys out of the model's context, so a prompt injection cannot read them.
• Vouch Shield a runtime check that inspects every tool call against your rules, like a customs officer at the door.
• Continuous trust heartbeats and session vouchers, so trust is a live signal that has to be renewed, not a badge that is issued once and trusted forever.
• Authority freshness a signed authority state with a counter that only goes up, so a state change (a fraud signal, a suspended mandate) collapses the acceptable freshness window to now for a high-consequence action, even when time-decay trust still passes. See docs/authority-freshness.md.

Framework integrations

Standalone packages that drop Vouch into the agent framework you already use. Each one issues a verifiable credential for a tool call, with optional delegation back to a human principal.

• vouch-langchain a LangChain tool that signs each tool call before it leaves the agent.
• vouch-langgraph signs LangGraph tool calls and graph nodes, so a signed trail runs across the whole graph.
• vouch-crewai a CrewAI tool, with supervisor-to-worker delegation that can only narrow authority, never widen it.
• vouch-a2a binds an A2A (Agent2Agent) Agent Card to a Vouch identity, so two agents can verify each other before they collaborate.
• vouch-goose registers the Vouch MCP server as an extension for Block's Goose agent, so a Goose session can sign and verify out of the box.
• vouch-mlflow signs an MLflow model artifact at registration time, bound to a content digest so any change to the weights breaks the signature.
• vouch-safetensors embeds a credential in a .safetensors header, complementary to OpenSSF Model Signing, so a model carries who produced it.

The standalone vouch-mcp package above ships alongside these.

SDKs, in the language you use

Python, TypeScript, and Go are the full reference implementations. A Rust core with idiomatic Swift, JVM (Java and Kotlin), .NET, and C wrappers shares one codebase, so every language produces byte-identical output, verified against shared test vectors. A WebAssembly build is included for the browser and the edge. See the table further down for status per language.

Robots and embodied agents

A robot is an agent with a body, so identity and accountability matter even more once it can cause physical harm. The vouch.robotics module ships six capabilities on the same eddsa-jcs-2022 credentials as the rest of Vouch: hardware-rooted identity (bound to a TPM or secure element, so it cannot be cloned to other hardware), model and config provenance (re-signable on every OTA update), physical capability scope (force, speed, a tighter cap near humans, allowed zones, and shift windows, checked before each actuation, with narrow-only delegation), a robot-to-robot trust handshake, an encrypted tamper-evident black box with a verifiable kill switch, and a scannable offline passport. All six are implemented in Python, TypeScript, Go, and the Rust core, which flows to the Swift, Kotlin/JVM, .NET, C/C++, and WebAssembly wrappers, so a robotics credential signed in one language verifies in every other. Fifteen further capabilities round out the set: a living-trust heartbeat (a signed per-interval motion summary whose trust decays unless it is renewed in-envelope), two-level credential revocation (per-credential status lists and whole-DID kill), an accountable safety record (a tamper-evident incident ledger summarized into a portable signed record), signed perception provenance (each captured sensor frame's hash bound to the robot's key and hash-linked, so a robot can prove what its sensors saw and a substituted frame is detectable), an offline delegation lease (a short-lived, scope-bounded grant a disconnected robot verifies and acts on with no network call, nesting across vendors), a physical quorum (a cryptographic two-person rule requiring M of N attested approvers for a high-consequence action), lifecycle credentials (ownership transfer that chains into a verifiable chain of custody, key rotation that forms a key history, and a signed decommission that retires the robot), a regulatory conformance profile (a machine-checkable mapping from a robot's credentials to the clauses of ISO 10218 and 15066, the EU Machinery Regulation, the EU AI Act, and UL 3300, with a deterministic checker and a signed conformance attestation), post-quantum signing by default (robot credentials carry a proof set with an Ed25519 proof and an ML-DSA-44 proof, with backward-compatible verification, so a robot stays unforgeable across its decade-long life), and cross-embodiment identity continuity (an agent identity that moves between robot bodies, with an embodiment credential re-binding to each body's hardware root and a continuity chain that proves the same accountable agent persisted, plus a fork check), a physical custody handoff chain (a signed record of who accepted custody of a task or object at each hop across human and robot actors, so an incident traces to the exact hop and an attested condition localizes damage to the holder responsible), bounded infrastructure access (an operator-signed grant naming a resource, its permitted operations, an optional zone, and a time window, which a robot pairs with a signed request the resource authorizes offline, so a robot can open a door or dock at a charger with a tamper-evident, attributable, shrink-only record), fused-sensor provenance (a signed attestation binding a robot's fused world model to the exact set of input frame hashes and the fusion method that produced it, so a manipulated fusion result or a dropped input is detectable and every fused input traces to a frame the robot recorded), wear and degradation attestation (a robot signs its own wear level bound to its identity and hash-linked over time, and a deterministic rule narrows its physical capability scope as it degrades, so a worn robot operates inside a tighter, verifiable envelope than the static limit it shipped with), and bystander-consent evidence (a robot binds the basis for capturing people, an explicit token, posted notice, a legitimate interest, or a redaction it applied, to the capture itself by its hash, holding only hashes and never anyone's identifying data, and a bystander can sign consent bound to one capture so it cannot be replayed to another), implemented across the same languages and pinned by the shared interop vector. See docs/robotics.md and the defensive disclosures PAD-064 through PAD-070, PAD-076 through PAD-084, and PAD-087 through PAD-094.

Inside your AI tools

• Claude Skill, OpenAI Custom GPT, and Gemini Gem packages that teach your AI assistant how to add Vouch to your code, running on your own AI subscription.

Media and the web

• C2PA Content Credentials for images.
• Vouch Sonic an audio watermark that carries provenance through sound.
• Browser extension for Chrome and Edge that signs and verifies content on the page.

For your repositories

• Gatekeeper GitHub App verifies commit signatures on every pull request and blocks leaked Vouch keys before they merge.

For the ecosystem

• Agent Trust Index an open benchmark that scans agents in the wild and measures how many can actually prove who they are. (Spoiler: today, almost none.)

Read the spec → | Join Discord →

---

The Problem

AI agents are making real-world API calls with ZERO cryptographic proof of:

• WHO they are
• WHAT they intended to do
• WHEN they did it

Examples of the risk:

• Healthcare AI accesses patient data → HIPAA violation risk
• Financial AI makes unauthorized trades → Liability nightmare
• Customer service AI leaks data → Compliance failure

Current solutions:

• DIY JWT signing → No agent-specific features, security mistakes easy
• Nothing → Most people just YOLO it and hope for the best

---

The Solution

Vouch Protocol provides cryptographic identity for AI agents, modeled after SSL/TLS:

• Verifiable Credentials (VC Data Model 2.0)
• Data Integrity proofs (eddsa-jcs-2022 cryptosuite, no JOSE/JWS dependency)
• Decentralized Identifiers (did:web, did:key)
• Multikey verification methods (algorithm-agnostic, post-quantum ready)
• Post-quantum profile (optional proof set: eddsa-jcs-2022 + mldsa44-jcs-2024)
• Human-readable JSON (proof attaches as a sibling object, no Base64-wrapped opaque payload)
• Framework-agnostic (works with MCP, LangChain, CrewAI, AutoGPT, AutoGen, Vertex AI)
• Cross-language interop (Python, TypeScript, Go, byte-identical canonical form)
• Open source (Apache 2.0 license, CC0 prior-art portfolio)

Think of it as:

• SSL certificate = Proves website identity
• Vouch Protocol = Proves AI agent identity

---

How It Works

The Workflow

flowchart LR
  P["👤 Principal<br/>did:web:user.example.com"]
  A["🤖 AI Agent<br/>did:web:agent.example.com<br/>+ Identity Sidecar"]
  C["📄 Vouch Credential<br/>VC + Data Integrity<br/>(eddsa-jcs-2022)"]
  API["🔐 API Endpoint"]
  V{"✅ Verified"}

  P -->|"Delegation VC"| A
  A -->|"sign(intent)"| C
  C -->|"HTTP body<br/>application/vc+vouch"| API
  API -->|"verify()"| V

4 Simple Steps:

1. Generate Identity: Create an Ed25519 keypair and a DID, publish a DID Document with a Multikey verification method.
2. Sign Action: Agent's sidecar issues a Verifiable Credential carrying action, target, and resource, secured by an eddsa-jcs-2022 Data Integrity proof.
3. Send to API: Transmit the credential as the HTTP request body with Content-Type: application/vc+vouch (or via the legacy Vouch-Token header for v0.x compatibility; the prior application/vouch+credential+json Content-Type is retained as a transitional alias).
4. Verify: API resolves the issuer's DID, validates the Data Integrity proof, checks temporal claims and the resource binding, returns a CredentialPassport.

The Trust Model

flowchart TB
  subgraph IDENTITY["Identity Layer"]
    DID["DID<br/>did:web / did:key"]
    MK["Multikey<br/>algorithm-agnostic key encoding"]
  end
  subgraph FORMAT["Credential Layer"]
    VC["Verifiable Credential<br/>(VC Data Model 2.0)"]
    INTENT["Intent payload<br/>action · target · resource"]
  end
  subgraph CRYPTO["Cryptographic Proof"]
    JCS["JCS canonicalization (RFC 8785)"]
    DEFAULT["eddsa-jcs-2022<br/>(Ed25519, default)"]
    PQ["mldsa44-jcs-2024<br/>(ML-DSA-44, optional second proof)"]
  end
  IDENTITY --> FORMAT
  FORMAT --> CRYPTO
  JCS --> DEFAULT
  JCS --> PQ

Trust = Verifiable Credentials + Data Integrity + Decentralized Identifiers + Multikey, with an optional post-quantum proof set. The same math that secures SSL/TLS, plus the standardized primitives that secure verifiable credentials elsewhere on the web, applied to AI agents.

---

Why Vouch Protocol?

vs. DIY JWT

Feature	Vouch Protocol	DIY JWT
Agent-specific	✅ (designed for agents)	❌ (generic)
MCP integration	✅ (native)	❌ (manual)
Framework integrations	✅ (LangChain, CrewAI, etc.)	❌
Audit trail format	✅ (VC standardized)	❌ (custom)
standards-aligned	✅ (eddsa-jcs-2022 Data Integrity)	❌
Multikey verification methods	✅ (algorithm-agnostic)	❌
Post-quantum signatures	✅ (mldsa44-jcs-2024 proof set)	❌
Cross-implementation interop tests	✅ (Python, TypeScript, Go)	❌
Security best practices	✅ (built-in)	⚠️ (easy to mess up)

---

Quick Start

1. Install

# Linux and macOS: one line
curl -fsSL https://vouch-protocol.com/install.sh | sh

# Or with pip (any platform)
pip install vouch-protocol

2. Generate Identity

vouch init --domain your-agent.com

3. Sign an Action (Agent Side)

v1.0 path (VC + Data Integrity, recommended):

from vouch import Signer
import os

signer = Signer(
  private_key=os.environ['VOUCH_PRIVATE_KEY'],
  did=os.environ['VOUCH_DID']
)

credential = signer.sign(intent={
  'action': 'read_database',
  'target': 'users_table',
  'resource': 'https://api.example.com/v1/users',
})
# Send credential as the JSON body of the API request, content-type
# application/vc+vouch

4. Verify (API Side)

v1.0 path:

from fastapi import FastAPI, Request, HTTPException
from vouch import Verifier

app = FastAPI()

@app.post("/api/resource")
async def protected_route(request: Request):
  credential = await request.json()
  public_key = '{"kty":"OKP", ...}' # Resolved from did:web or trusted root

  is_valid, passport = Verifier.verify(credential, public_key=public_key)
  if not is_valid:
    raise HTTPException(status_code=401, detail="Untrusted Agent")

  return {
    "status": "Verified",
    "agent": passport.sub,
    "intent": passport.intent,
  }

That's it. A few lines to sign, a few to verify, on either path.

---

Integrations

Works with all major AI frameworks out-of-the-box:

• Model Context Protocol (MCP): native integration for Claude Desktop and Cursor
• LangChain: sign tool calls automatically
• CrewAI: multi-agent identity management
• AutoGPT: autonomous agent signing
• AutoGen: Microsoft multi-agent framework
• Google Vertex AI: sign function calls
• Google ADK: native ADK tool integration
• n8n: low-code agent workflows

See all integrations →

---

Enterprise Features

• Key Rotation: automatic rotating keys for production
• Voice AI Signing: sign audio frames in real-time
• Cloud KMS: AWS KMS, GCP Cloud KMS, Azure Key Vault
• Reputation Scoring: track agent behavior over time
• Revocation Registry: blacklist compromised keys
• Redis Caching: production-scale verification
• Post-Quantum Profile: an optional proof set carrying an eddsa-jcs-2022 proof and an mldsa44-jcs-2024 proof for regulated deployments aligning with NIST CNSA 2.0 / NSM-10 migration timelines

Post-Quantum Example

# Optional profile, requires `pip install pqcrypto`
credential = signer.sign_hybrid(intent={
  'action': 'submit_clinical_finding',
  'target': 'trial:NCT00000001',
  'resource': 'https://fda-submissions.example.com/api/findings',
})
# The credential's `proof` is an array of two independent Data Integrity proofs,
# eddsa-jcs-2022 and mldsa44-jcs-2024, over the same document.
# Verification REQUIRES both to validate.

---

Use Cases

Financial Services

# SEC-compliant trade logging
agent.sign({'action': 'execute_trade', 'amount': 10000, 'symbol': 'AAPL'})

Customer Service

# Data access accountability
agent.sign({'action': 'read_customer_data', 'customer_id': 'cust_abc'})

Healthcare AI

# HIPAA-compliant audit trail
agent.sign({'action': 'access_phi', 'patient_id': '12345'})

To keep this space free from patent capture, the project publishes 61 defensive prior-art disclosures under CC0, covering cryptographic identity, media provenance, voice biometrics, AI safety, post-quantum cryptography, AI coding governance, and per-region human-or-AI code authorship. See docs/disclosures.

---

Use cases

• Financial services. A signed, accountable record of every trade or transfer an agent makes.
• Healthcare. An auditable trail for every access to patient data.
• Customer service. Proof of which agent touched which customer record, and on whose authority.
• Agent to agent. When one organization's agent calls another's, each can verify the other before acting.

See full examples →

---

Documentation

• Quick Start
• Vouch Specification (v1.6 normative)
• Post-Quantum Implementation Guide
• Protocol Specification (developer guide)
• Integration Guides
• Threat Model and Security Considerations
• Defensive Prior Art Disclosures (61 PADs, CC0)
• FAQ and Discussions

---

Community

• Discord - Ask questions, share use cases → Join now
• GitHub Discussions - Technical discussions → Start a discussion
• Twitter/X - Updates and announcements → @Vouch_Protocol

---

Roadmap