Skip to main content
ClaudeWave
MCP ServersOfficial Registry0 stars0 forks● JavaScriptMITUpdated today
ClaudeWave Trust Score
77/100
✓ Trusted
Passed
  • ✓Open-source license (MIT)
  • ✓Actively maintained (<30d)
  • ✓Documented (README)
Flags
  • !No description
Last scanned: 10/15/2026
Install in Claude Code / Claude Desktop
Method: Manual
Claude Code CLI
git clone https://github.com/cdonoyan/shuka-agent
claude_desktop_config.json (Claude Desktop)
{
  "mcpServers": {
    "shuka-agent": {
      "command": "node",
      "args": ["/path/to/shuka-agent/dist/index.js"]
    }
  }
}
1. Run the command above in your terminal (Claude Code), or paste the JSON config into claude_desktop_config.json (Claude Desktop).
2. Replace any <placeholder> values with your API keys or paths.
3. Restart Claude. The MCP server and its tools appear automatically.
💡 Clone https://github.com/cdonoyan/shuka-agent and follow its README for install instructions.
Use cases

MCP Servers overview

# @shukashake/agent

**Universal Trust Guardian for AI Agents**

The Shuka Shake is the trust layer for moving protected data in the AI economy. It doesn't own what's inside - it guards the exchange, respects confidentiality, and enforces terms without ever touching the metadata within.

This package is your agent's membership to the yacht club - where trust is negotiated, proof tokens are exchanged, and data flows with respect for rights.

**v2.1.0 — Everything is a Shuka Shake.** This is the canonical One Ring package: one SDK and one MCP server uniting attestation, the rules layer, the shake-anchored license marketplace, enterprise licensing, and the registry ledger. The full flow:

```
shake → terms (rules layer) → license offering → grant → registry proof
```

1. **Shake** — create an idempotent Shuka Shake for any entity (same input, same shake, forever)
2. **Terms** — attach modifiable rules to the immutable handshake
3. **Offering** — issue a license offering anchored to the shake
4. **Grant** — purchase the offering; receive a grant with its own proof token
5. **Registry proof** — verify the grant and its full ledger chain, publicly

---

## ⚠️ Partner Access Required

**This SDK requires a verified partner account and an SDK key.**

Authentication is SDK-key-only: every authenticated operation uses your SDK key (sent as `X-SDK-API-Key`). Most operations route through Shuka's `/api/sdk/v1/*` surface; verification and trust-negotiation use the public `/api/shake/*` and `/api/handshake/*` endpoints (verification is free and needs no key). Agents and developers never handle JWTs or raw platform API keys. Scoped permissions (e.g. `license:read`, `license:write`, `transaction:write`) are attached to the key at issuance.

You can install this package, but it will not function without an authorized SDK key issued by Auron.

**To become a partner:**
- Email [info@auron.co](mailto:info@auron.co)
- Connect with leadership via [shuka.app](https://shuka.app)

Once approved, you'll receive API credentials to activate the SDK.

**Attribution model:** the key's user binding IS the attribution. Every shake
this SDK mints is owned by the partner user your key is assigned to
(`user_id` / `partner_id`) and stamped with the minting key (`api_key_id`).
Per-end-user attribution inside your app is your client reference (the `id`
field on attest calls), which lands in the idempotency namespace and
`handshake_metadata`. Agent identity (`X-Agent-Id` / `X-Agent-Type`) is
recorded as labeled-untrusted `asserted_agent` metadata for audit — it never
sets ownership. If your key is unassigned, creations degrade to anonymous
(`"access": "public"`); claim the key via `POST /api/sdk/v1/partner/keys/claim`.

---

## What is Shuka?

Shuka provides cryptographically verified trust handshakes for AI-to-AI and AI-to-human data transfers. Every exchange is attested, auditable, and verifiable.

**Use cases:**
- Healthcare ROI (Release of Information) management
- Construction contractor/permit verification
- Cross-jurisdiction compliance attestation
- Data portability between enterprise systems
- AI agent verification and trust negotiation

**The Shuka Shake is:**
- **Schema-agnostic** - any data format works
- **Industry-agnostic** - healthcare, construction, finance, legal, government
- **Jurisdiction-agnostic** - works across borders and regulatory frameworks

## Install

```bash
npm install @shukashake/agent
```

## Quick Start

```javascript
const { ShukaSDK } = require('@shukashake/agent');

const shuka = new ShukaSDK({
  apiKey: process.env.SHUKA_API_KEY  // your SDK key
});

// Create an attestation (the Shuka Shake)
const result = await shuka.attest(
  'Patient John Doe consents to release records to Dr. Smith',
  { industry: 'healthcare', jurisdiction: 'US' }
);

// Share the proof token - portable, verifiable, universal
console.log(result.proof_token); // shk_v1.xxx.yyy
```

### Quick Start: The United Flow

```javascript
// 1. SHAKE — idempotent: same entity data always yields the same shake
const shake = await shuka.shake('brand', {
  name: 'ACME Studios',
  registration: 'US-TM-12345'
}, { purpose: 'brand identity attestation', industry: 'universal' });

// 2. TERMS — attach modifiable rules to the immutable handshake
await shuka.createTerms(shake.handshake_id, {
  accessType: 'time_limited',
  permittedPurposes: ['licensing', 'verification']
});

// 3. OFFERING — issue a license offering anchored to the shake
const offering = await shuka.issueLicense({
  anchorHandshakeId: shake.handshake_id,
  brand: 'ACME Studios',
  rightsHolder: 'ACME Holdings LLC',
  validUntil: '2027-12-31T23:59:59Z',
  buyToOpenPrice: 500
});

// 4. GRANT — purchase the offering (buyer side)
const grant = await shuka.purchaseLicense(offering.license.id, 500);

// 5. REGISTRY PROOF — verify the grant and its ledger chain
const proof = await shuka.verifyGrant(grant.grant_id);
```

## Core Concepts

### Proof Tokens

When you create an attestation, Shuka returns a proof token:

```
shk_v1.YWJjMTIzZGVm.a1b2c3d4
```

This token is:
- **Portable** - share it anywhere, verify it anywhere
- **Verifiable** - anyone with API access can verify it
- **Persistent** - the attestation record is immutable

### Trust Scores

Every attestation has a trust score (0.0 - 1.0):
- **0.8+** - Safe to proceed
- **0.6-0.8** - Proceed with caution
- **Below 0.6** - Additional verification recommended

### Handshake Types

Shuka supports multiple handshake types for different use cases:

| Type | Use Case |
|------|----------|
| `attestation` | General attestations (default) |
| `license` | License verification |
| `compliance` | Compliance attestation |
| `compliance_attestation` | Compliance with attestation |
| `release_of_information` | Healthcare ROI |
| `partner_compliance` | Partner compliance verification |

## API Reference

### ShukaSDK

Main SDK class for AI agent integration.

```javascript
const { ShukaSDK } = require('@shukashake/agent');

const shuka = new ShukaSDK({
  apiKey: 'your-api-key',
  agentName: 'my-agent'  // For audit trails
});
```

#### attest(description, options?)

Create a verifiable attestation.

```javascript
const result = await shuka.attest(
  'Contractor ABC Corp is licensed for electrical work in Texas',
  {
    industry: 'construction',
    jurisdiction: 'US-TX',
    metadata: {
      contractor_id: 'ABC-123',
      license_number: 'TX-ELEC-456'
    }
  }
);

// Returns:
// {
//   success: true,
//   proof_token: 'shk_v1.xxx.yyy',
//   handshake_id: 'uuid',
//   trust_score: 0.85
// }
```

#### verify(proofToken)

Verify a proof token.

```javascript
const result = await shuka.verify('shk_v1.xxx.yyy');

// Returns:
// {
//   verified: true,
//   trust_score: 0.85,
//   handshake_id: 'uuid',
//   created_at: '2024-01-01T00:00:00Z'
// }
```

#### negotiate(proofToken, purpose?)

Agent-to-agent trust negotiation. When you receive data with a proof from another agent, ask Shuka to vouch for their attestation.

```javascript
const trust = await shuka.negotiate(
  incomingProofToken,
  'processing patient data for care coordination'
);

if (trust.recommendation === 'safe_to_proceed') {
  // The data exchange is trusted - proceed
} else if (trust.recommendation === 'proceed_with_caution') {
  // Consider additional verification
} else {
  // Do not proceed
}
```

#### acknowledgeReceipt(envelopeId)

Complete the exchange by acknowledging receipt (the S'more model - both ends warm).

```javascript
await shuka.acknowledgeReceipt('ENV-xxx');
```

### Terms and Conditions

The terms layer is how data owners set constraints on how their data can be used, and how recipients understand and accept those constraints.

#### Creating Attestations with Terms

Attach terms when creating attestations to control how data can be accessed:

```javascript
const result = await shuka.attest(
  'Patient John Doe (US, healthcare) consents to release records to Dr. Smith',
  {
    terms: {
      accessType: 'time_limited',           // How data can be accessed
      validityDays: 365,                    // Days the terms remain valid
      requireReceipt: true,                 // Accesses must be receipted
      permittedPurposes: ['care_coordination', 'treatment'],
      prohibitedPurposes: ['marketing', 'research']
    }
  }
);
```

Include industry/jurisdiction context in the attested text itself — the
quick-attest endpoint records the text and terms. For richer constraints
(max uses, downstream rules, pricing), attach full terms afterwards with
`createTerms(handshakeId, terms)` (see Rules Layer below).

#### Access Types

| Type | Description |
|------|-------------|
| `single_use` | One-time access, then terms expire |
| `time_limited` | Access valid for a time period |
| `use_limited` | Access valid for N uses |
| `perpetual` | Unlimited access (rare, high trust) |
| `session_bound` | Access tied to active session |
| `windowed` | Access only during specific time windows |

#### getTerms(handshakeId)

Get the terms attached to a handshake before processing its data:

```javascript
const result = await shuka.getTerms('handshake-uuid');

console.log(result.terms.access_type);         // 'time_limited'
console.log(result.terms.permitted_purposes);  // ['care_coordination', 'treatment']
console.log(result.terms.id);                  // terms_id for the ID-based methods
```

> **Removed in 2.1.1:** `acceptTerms`, `checkAccess`, `revokeAccess`,
> `updateTerms`, `getTermsHistory`, and `getTermsAcceptors` called legacy
> routes that never existed server-side (every call 404'd). They now throw
> `NOT_SUPPORTED` with a pointer to the working equivalent in the Rules
> Layer: `recordTermsAccess`, `revokeTermsById`, `updateTermsById`,
> `getTermsRevisions`, and `getTermsReceipts` — all keyed by the `terms_id`
> returned from `getTerms(handshakeId)` or `createTerms()`.

#### getChainOfTrust(handshakeId)

Get the full provenance chain for an attestation.

```javascript
const chain

What people ask about shuka-agent

What is cdonoyan/shuka-agent?

+

cdonoyan/shuka-agent is mcp servers for the Claude AI ecosystem with 0 GitHub stars.

How do I install shuka-agent?

+

You can install shuka-agent by cloning the repository (https://github.com/cdonoyan/shuka-agent) or following the README instructions on GitHub. ClaudeWave also provides quick install blocks on this page.

Is cdonoyan/shuka-agent safe to use?

+

Our security agent has analyzed cdonoyan/shuka-agent and assigned a Trust Score of 77/100 (tier: Trusted). See the full breakdown of passed checks and flags on this page.

Who maintains cdonoyan/shuka-agent?

+

cdonoyan/shuka-agent is maintained by cdonoyan. The last recorded GitHub activity is dated 2026-10-14, with 0 open issues.

Are there alternatives to shuka-agent?

+

Yes. On ClaudeWave you can browse similar mcp servers at /categories/mcp, sorted by popularity or recent activity.

Deploy shuka-agent to your cloud

Ship this repo to production in minutes. Each platform spins up its own environment with editable env vars.

Maintain this repo? Add a badge to your README

Drop the badge into your GitHub README to show it's tracked on ClaudeWave. Each badge links back to this page and reflects the live Trust Score.

Featured on ClaudeWave: cdonoyan/shuka-agent
[![Featured on ClaudeWave](https://claudewave.com/api/badge/cdonoyan-shuka-agent)](https://claudewave.com/repo/cdonoyan-shuka-agent)
<a href="https://claudewave.com/repo/cdonoyan-shuka-agent"><img src="https://claudewave.com/api/badge/cdonoyan-shuka-agent" alt="Featured on ClaudeWave: cdonoyan/shuka-agent" width="320" height="64" /></a>

More MCP Servers

shuka-agent alternatives