Skip to content
MCP server

Prescan

Rule-based scans for prompt injection, poisoned MCP tools and unsafe x402 endpoints

Install
Pricing
Paid
Platforms
API
Runs
Remotely
Version
0.1.0
Listing updated
2 days ago

About Prescan

Prescan is a paid MCP server for security. It runs as a hosted remote server, so there is nothing to install locally.

Prescan runs deterministic checks that an agent calls before acting on untrusted input. No language model is involved, scans are described as sub-millisecond, every finding has a span, and the rules and their published precision and recall can be read through the API.

The text scan looks for instruction override in eight languages, role hijack, jailbreaks, system-prompt extraction, data exfiltration, tool or fund hijack, crypto lures, hidden Unicode and CSS text, and base64, hex or percent-encoded payloads that it decodes and scans again. It returns a safe, suspicious or malicious verdict, a 0 to 100 score, a sanitised copy and takes up to 100 texts per batch. The tool scan inspects a tools/list result for hidden instructions, cross-tool shadowing and lookalike names, and flags capabilities such as shell, files, network and payments. The x402 check fetches a payment endpoint's terms without paying and returns go, caution or no_go.

The paid tools cost $0.002 to $0.003 per call in USDC on Base via x402, charged only on success. The disclaimer says a clean result lowers risk but does not prove content is safe, and inputs are not stored.

Install

Run in a terminal

claude mcp add --transport http prescan https://prescan.canisell.xyz/mcp

Add to Settings → Connectors → Add custom connector

https://prescan.canisell.xyz/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

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

Add to .vscode/mcp.json

{
  "servers": {
    "prescan": {
      "type": "http",
      "url": "https://prescan.canisell.xyz/mcp"
    }
  }
}

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

{
  "mcpServers": {
    "prescan": {
      "serverUrl": "https://prescan.canisell.xyz/mcp"
    }
  }
}

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

{
  "mcpServers": {
    "prescan": {
      "type": "http",
      "url": "https://prescan.canisell.xyz/mcp"
    }
  }
}

Tools this server provides (5)

Taken from the project documentation; the running server may differ.

scan_text
Prompt-injection scan of text, single or batch
scan_tools
Tool-poisoning scan of MCP tool definitions
check_x402_endpoint
Check whether an x402 endpoint is safe to pay
preview_scan_text
Free preview scan of up to 1,000 characters
list_rules
List every rule and scoring method

What it does

Prompt-injection scan
Verdict, score, findings with character spans and a sanitised copy.
Tool-poisoning scan
Checks MCP tool definitions for hidden instructions and shadowing.
x402 endpoint check
Returns go, caution or no_go without paying the endpoint.
Published rules
Every rule and the scoring method are readable via the API.
Free preview
Free tools preview text up to 1,000 characters and list rules.
Batch scanning
Up to 100 texts per call.

Good for

  • Screen a fetched web page before an agent reads it
  • Audit an MCP server's tool list before connecting
  • Check an x402 seller before paying
  • Sanitise tool output that contains hidden text

Pricing

Paid

$0.002 to $0.003 per call via x402; free preview and rules

Prices change. Check prescan.canisell.xyz before you buy.

Prescan FAQ

How much does it cost?

The text and tool scans are $0.002 each and the x402 check is $0.003, paid in USDC on Base, charged only on success. A preview and the rule list are free.

Does it use an LLM?

No. The checks are deterministic rules.

Does it guarantee safety?

No. The disclaimer says results are heuristic signals and a clean result does not prove content is safe.

Are inputs stored?

The service states that inputs are not stored.

Which tools are exposed?

scan_text, scan_tools, check_x402_endpoint, preview_scan_text and list_rules.

Is Prescan free?

No. Prescan is a paid product.

How do I add Prescan to Claude Code?

Run this command in your terminal: claude mcp add --transport http prescan https://prescan.canisell.xyz/mcp

What are the best alternatives to Prescan?

The closest alternatives to Prescan are REA, Agent Roast, Midplane, Leash and Tinplate Edge AI Safety, Security & EU Compliance.

Alternatives to Prescan

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Agent Roast

    MCP server

    Offline review that flags agent changes which weaken tests, assertions or CI

    Runs locally
    • Open source
    • 4 Hacker News points
    • Developer tools
  3. Midplane

    MCP server

    Postgres gateway for AI agents with table grants, column masking and approval for writes

    Runs locally
    • Open source
    • 5 GitHub stars
    • Databases
  4. Leash

    MCP server

    Review what your AI touched, approve risky actions and undo supported changes

    • 1 tool
    • Remote
    • Free
    • 1 GitHub star
    • Security
  5. SecureCloud

    MCP server

    SentinelOne and IT security catalog, prices, quotes and checkout links for Israel

    • 7 tools
    • Remote
    • Free
    • Security