
by 1clawAI
1Claw MCP gives AI agents just-in-time access to secrets stored in a 1Claw vault, so keys are used without being pasted into prompts. It also includes tools for transaction signing and a security scanner for LLM content.
PluginContact for pricing
Step 1: Run npx @1claw/cli setup to log in and create an agent, vault and policy.
Step 2: Or add a 1claw entry to your MCP config with command npx and args -y @1claw/mcp.
Step 3: Set ONECLAW_AGENT_API_KEY in the entry's env.
Step 4: For a hosted setup, point a client that supports HTTP streaming at mcp.1claw.co/mcp.
Step 5: Set ONECLAW_LOCAL_ONLY=true if you only want the security scanner.
No. Credentials stay server-side and agents run requests through bindings without seeing the keys. Access uses short-lived JWTs with scoped policies, and actions go to an audit log.
Yes, in local-only mode the inspect_content tool runs without vault credentials.
The server has 162 tools, but each session sees only its entitled toolsets. A vault-only agent sees 28.
Requirements
Aikido SecurityAikido Security is a platform that scans code, cloud, containers and running apps for vulnerabilities, then opens pull requests with fixes. It also offers autonomous pentesting and device and runtime protection.
Snyk DeepCode AISnyk is a security platform that scans the code, dependencies, containers and infrastructure configuration your team ships, including code written by AI. It also covers AI coding agents and AI-native applications.