Atlas / MCP servers / kasarlabs / Snak

SnakSAFE

Name: Snak
Author: kasarlabs

mcp/kasarlabs/snak

MCP server by kasarlabs · Audited 2026-10-07

⚑ Report / re-scan Source ↗

Build powerful and secure AI Agents powered by Starknet.

Verdict

SAFE

Grade

Trust score

89 /100

Exposed tools

23 16r · 5w · 2d

Transport

—

License

MIT

Stars

Overview

From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.

A Agent Engine for creating powerful and secure AI Agents powered by Starknet. Available as both an NPM package and a ready-to-use backend.

Quick Start

Prerequisites

Starknet wallet (recommended: Argent X)
AI provider API key (Anthropic/OpenAI/Google Gemini/Ollama)
Docker and Docker compose installed
Node.js and pnpm installed

Installation

git clone https://github.com/kasarlabs/snak.git
cd snak
pnpm install

Configuration

Create a .env file by copying .env.example:

cp .env.example .env

Then, fill in the necessary values in your .env file:

# --- Starknet configuration (mandatory) ---
STARKNET_PUBLIC_ADDRESS="YOUR_STARKNET_PUBLIC_ADDRESS"
STARKNET_PRIVATE_KEY="YOUR_STARKNET_PRIVATE_KEY"
STARKNET_RPC_URL="YOUR_STARKNET_RPC_URL"

# --- AI Model API Keys (mandatory) ---
# Add the API keys for the specific AI providers you use in config/models/default.models.json
# T

Read from source at commit 608b1798e8c4OBSERVED · 2026-10-07

Connect

Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.

claude-code

claude mcp add test --env ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY} --env DEFAULT_MAX_TOKENS=${DEFAULT_MAX_TOKENS} --env GEMINI_API_KEY=${GEMINI_API_KEY} --env OPENAI_API_KEY=${OPENAI_API_KEY} -- npx -y @snakagent/[email protected]

claude-desktop

{
  "mcpServers": {
    "test": {
      "command": "npx",
      "args": [
        "-y",
        "@snakagent/[email protected]"
      ],
      "env": {
        "ANTHROPIC_API_KEY": "${ANTHROPIC_API_KEY}",
        "DEFAULT_MAX_TOKENS": "${DEFAULT_MAX_TOKENS}",
        "GEMINI_API_KEY": "${GEMINI_API_KEY}",
        "OPENAI_API_KEY": "${OPENAI_API_KEY}"
      }
    }
  }
}

Exposed tools (23)

16 read · 5 write · 2 destructive. Blast radius: 2 tools can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.

Tool	Risk	Description
`Test_Integration_Agent`	read	Full integration test agent
`add_mcp_server`	write	Add an MCP server
`agent-special`	read	Handles special operations
`agent1`	read	Handles blockchain operations
`agent2`	read	Handles configuration management
`agent3`	read	Handles MCP operations
`agent4`	read	Handles MCP operations for user2
`ask_human`	read	[SNAK CORE TOOL] Use this tool to ask the user for input when necessary don
`block_task`	read	[SNAK CORE TOOL] Use when the task cannot be completed due to unresolvable obstacles. Provide details in the response.
`create_agent`	write	Create a new agent configuration with default settings. Only accepts agent profile (name, group, description, contexts). All other configuration (graph, memory, rag, mcp_servers) will use default values. To modify configuration after creation, use update_agent tool.
`create_task`	write	[SNAK Tool] Create a structured task with thoughts, reasoning, criticism, and speak fields
`delete_agent`	destructive	Delete/remove/destroy an agent configuration permanently. Use when user wants to delete, remove, or destroy an agent completely.
`end_task`	read	[SNAK CORE TOOL] End the current task successfully. Provide details in the response.
`list_agents`	read	List/show/get all agent configurations with optional filtering. Use when user wants to see multiple agents, all agents, or find agents matching certain criteria.
`message_ask_user`	read	Ask user a question and wait for response. Use for requesting clarification, asking for confirmation, or gathering additional information. This tool creates an interrupt in the graph execution that pauses until the user provides a response.
`read_agent`	read	Get/retrieve/show/view/find details and configuration of a specific agent by ID or name. Use when user wants to see information about a particular agent.
`remove_mcp_server`	destructive	Remove/uninstall/delete one or multiple MCP servers from an agent. Use when user wants to remove MCP server capabilities from an existing agent.
`response_task`	read	[SNAK Tool] Provide a structured response with thoughts, reasoning, criticism, and speak fields
`search_mcp_server`	read	Search for MCP servers
`test-agent`	read	Test agent
`tool1`	read	Test tool
`update_agent`	write	Update/modify/change/rename specific properties of an existing agent configuration. Use when user wants to modify, change, update, edit, or rename any agent property like name, description, group, etc.
`update_mcp_server`	write	Update/modify/change configuration of one or multiple existing MCP servers for a specific agent. Use when user wants to modify MCP server settings or configuration.

Trust audit

SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.

Layer	What it checks	Result
L0	Provenance & inventory	PASS
L1	Static analysis of the code	PASS
L2	Instruction surface (what it tells the agent)	PASS
L3	Class-specific surface	WARN
L4	Behavioural (sandbox)	SKIPPED

What the source does

Filesystem: declared (4 observation(s))
Network: declared (5 observation(s))
Shell: none-observed
Dependencies: not all pinned
Secrets in source: none-found

Findings (15)

MEDIUMFilesystem / path · mcp.destructive_tools · CWE-22, CWE-59

delete_agent, remove_mcp_server

Why it matters. 2 tool(s) can delete or overwrite

Fix. prefer a read-only mode or scoped tokens; the page states the blast radius

LOWInventory / provenance · inv.hidden_file · CWE-1104

.node-version

.node-version

Why it matters. hidden member outside the usual dotfiles

Fix. review its purpose

LOWInventory / provenance · inv.hidden_file · CWE-1104

.prettierignore

.prettierignore

Why it matters. hidden member outside the usual dotfiles

Fix. review its purpose

LOWInventory / provenance · inv.hidden_file · CWE-1104

.prettierrc.json

.prettierrc.json

Why it matters. hidden member outside the usual dotfiles

Fix. review its purpose

LOWFilesystem / path · fs.traversal · CWE-22, CWE-59

packages/agent/jest.config.cjs:28

'^../../config/agentConfig\\.js$': '<rootDir>/src/__mocks__/agentConfig.ts',

LOWFilesystem / path · fs.traversal · CWE-22, CWE-59

packages/agent/src/agents/core/__tests__/utils.spec.ts:11

jest.mock('../../../tools/tools.js');

LOWFilesystem / path · fs.traversal · CWE-22, CWE-59

packages/agent/src/agents/core/__tests__/utils.spec.ts:12

jest.mock('../../../services/mcp/src/mcp.js');

LOWFilesystem / path · fs.traversal · CWE-22, CWE-59

packages/agent/src/agents/core/__tests__/utils.spec.ts:18

require('../../../tools/tools.js').createAllowedTools

LOWFilesystem / path · fs.traversal · CWE-22, CWE-59

packages/agent/src/agents/core/__tests__/utils.spec.ts:21

require('../../../services/mcp/src/mcp.js').MCP_CONTROLLER

LOWSupply chain · supply.unpinned · CWE-829, CWE-1357

package.json

@langchain/anthropic, @langchain/community, @langchain/core, @langchain/deepseek, @langchain/google-genai, @langchain/langgraph, @langchain/langgraph-checkpoint, @langchain/langgraph-checkpoint-postgr

Why it matters. 61 dependency range(s) float