Skip to content
MCP server

Caelara Labs Agent Safety

Remote MCP server with agent safety tools, pay-per-call via x402 USDC or API key

Answered our latest tool-list check 6 hours ago with 10 tools. It will be marked verified after 3 successful checks in 7 days.

Install
Pricing
Paid
Platforms
API
Runs
Remotely
Version
1.0.1
Listing updated
1 day ago

Similar MCP servers

About Caelara Labs Agent Safety

Caelara Labs Agent Safety is a paid MCP server for agent tooling. It runs as a hosted remote server, so there is nothing to install locally.

The website returns only a status payload listing an OpenAPI description, an MCP endpoint, an x402 payment manifest, an A2A endpoint and an agent card, so the public page itself says little about the individual safety checks offered. The registry entry describes a remote streamable-HTTP MCP server named for Caelara Labs that exposes agent safety tools to MCP-compatible clients.

Access is metered: according to the registry entry, each call can be paid per request in USDC over x402, or made with an API key. The same host also publishes an A2A endpoint and agent card, which suggests the service can be reached by other agents as well as by MCP clients.

The listing is most relevant to developers who want to add safety or guardrail checks to an AI agent workflow and prefer pay-as-you-go access over a subscription. Specific tool names and pricing amounts are not visible in the material available here, so check the OpenAPI document or MCP tool list before integrating.

Install

Run in a terminal

claude mcp add --transport http agent https://agent.caelaralabs.com/mcp

Add to Settings → Connectors → Add custom connector

https://agent.caelaralabs.com/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "agent": {
      "url": "https://agent.caelaralabs.com/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "agent": {
      "type": "http",
      "url": "https://agent.caelaralabs.com/mcp"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "agent": {
      "serverUrl": "https://agent.caelaralabs.com/mcp"
    }
  }
}

Add to Grok Bot desktop app → Connect apps → Add a custom connector

https://agent.caelaralabs.com/mcp

Enter a Name (for example "Caelara Labs Agent Safety"), choose Type: Remote HTTPS, paste this as the Server URL and choose Add plugin.

Full Grok Bot setup guide

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "agent": {
      "type": "http",
      "url": "https://agent.caelaralabs.com/mcp"
    }
  }
}

Using several servers? Build one config file with the MCP config generator.

Tools this server provides (10)

Read directly from the server.

agent_input_firewall
Scan untrusted input: injection risk + PII redaction + sanitized text + allow/review/block verdict
agent_output_validator
Validate LLM output: repair JSON, enforce schema, check grounding against a source, extract citations
pii_redactor
Redact emails, phones, SSNs, cards (Luhn), IPs, API keys and JWTs
prompt_injection_scanner
Score text for prompt-injection patterns
token_compressor
Compress text to a target ratio of its original length
json_auto_repair
Repair malformed JSON from LLM output
schema_enforcer
Validate and coerce data to a JSON Schema
hallucination_diff
Flag answer sentences unsupported by the source text
citation_extractor
Extract URLs, DOIs, arXiv ids and author-year citations
context_chunker
Split text into token-bounded overlapping chunks

What it does

Remote MCP endpoint
Served over streamable HTTP at a hosted URL, so clients connect without installing a local package.
Pay-per-call access
Calls can be paid per request in USDC via x402, as described in the registry entry.
API key option
The registry entry also lists API key authentication as an alternative to x402 payment.
Multiple agent interfaces
The host advertises an OpenAPI spec, an A2A endpoint and an agent card alongside MCP.

Good for

  • Adding safety checks to an MCP-connected AI agent
  • Paying per call for agent safety tooling without a subscription
  • Exposing safety endpoints to other agents over A2A
  • Discovering the service's OpenAPI definition for direct API integration

Pricing

Paid

Pay-per-call via x402 USDC or API key; per-call amounts not stated.

Prices change. Check agent.caelaralabs.com before you buy.

Caelara Labs Agent Safety FAQ

How is the server accessed?

The registry entry lists a remote streamable-HTTP endpoint at https://agent.caelaralabs.com/mcp.

How can I pay for calls?

The registry entry says calls are pay-per-call via x402 USDC or with an API key.

Is it open source?

Nothing in the provided material mentions a source repository or open-source licence.

Is Caelara Labs Agent Safety free?

No. Caelara Labs Agent Safety is a paid product.

How do I add Caelara Labs Agent Safety to Claude Code?

Run this command in your terminal: claude mcp add --transport http agent https://agent.caelaralabs.com/mcp

What are the best alternatives to Caelara Labs Agent Safety?

The closest alternatives to Caelara Labs Agent Safety are Caelara Labs Web Intelligence, Caelara Labs Compliance & Trust, Caelara Labs On-Chain Intelligence, Caelara Labs Document Utilities and Caelara Labs Social Analytics.

Where this comes from

  • Description: written from the project's website and its MCP Registry entry.
  • Status and tools: checked automatically by AIPoweredBot on 20 November 2026.
  • Install steps: from the official MCP Registry entry.

How AIPoweredBot checks servers