Skip to main content
Glama
guardd

Orcho MCP Server

by guardd

Orcho MCP Server for Cursor

Risk assessment for AI coding prompts - Automatically analyze your coding requests for security and safety risks before execution.

šŸš€ Quick Install

Click this link to automatically install in Cursor:

cursor://anysphere.cursor-deeplink/mcp/install?name=orcho&config=eyJuYW1lIjoib3JjaG8iLCJ0eXBlIjoic3RkaW8iLCJjb21tYW5kIjoibnB4IiwiYXJncyI6WyIteSIsIkBvcmNob19yaXNrL21jcC1zZXJ2ZXIiXSwiZW52Ijp7Ik9SQ0hPX0FQSV9LRVkiOiJ0ZXN0X2tleV9vcmNob18xMjM0NSJ9fQ==

How to use:

  1. Copy the link above

  2. Paste it into your browser's address bar and press Enter

  3. Cursor will open and automatically configure the MCP server

  4. Replace the test API key with your real key (see API Configuration)

  5. Restart Cursor to activate

  6. Replace the test API key with your real key (see API Configuration)

  7. Restart Cursor to activate the MCP server


Related MCP server: Sekrd Security Scanner

What is Orcho??

Orcho analyzes your coding prompts in real-time to identify potential security risks, dangerous operations, and safety concerns before code is generated or executed.

Features

  • šŸ” Real-time Risk Assessment - Analyze prompts using Orcho's risk analysis API

  • šŸ“ Context-Aware - Automatically includes file context for accurate blast radius and complexity analysis

  • šŸ›”ļø Security First - Identifies high-risk prompts before execution

  • šŸ”Œ Seamless Integration - Works natively with Cursor's Model Context Protocol


Installation

Copy and paste this link into your browser:

cursor://anysphere.cursor-deeplink/mcp/install?name=orcho&config=eyJuYW1lIjoib3JjaG8iLCJ0eXBlIjoic3RkaW8iLCJjb21tYW5kIjoibnB4IiwiYXJncyI6WyIteSIsIkBvcmNob19yaXNrL21jcC1zZXJ2ZXIiXSwiZW52Ijp7Ik9SQ0hPX0FQSV9LRVkiOiJ0ZXN0X2tleV9vcmNob18xMjM0NSJ9fQ==

This automatically:

  • āœ… Configures the MCP server in Cursor

  • āœ… Sets up auto-installation via npx

  • āš ļø Next step: Replace the test API key with your real key (see below)

  • āš ļø Then: Restart Cursor

Option 2: Manual Installation

  1. Install the package:

    npm install -g @orcho_risk/mcp-server
  2. Configure Cursor:

    Create or edit ~/.cursor/mcp.json (Windows: C:\Users\<YourUsername>\.cursor\mcp.json):

    {
      "mcpServers": {
        "orcho": {
          "command": "npx",
          "args": ["-y", "@orcho_risk/mcp-server"],
          "env": {
            "ORCHO_API_KEY": "your-api-key-here"
          }
        }
      }
    }
  3. Restart Cursor completely (quit and reopen)


API Configuration

Get Your API Key

  1. Sign up at app.orcho.ai

  2. Navigate to API Settings (Dashboard → API Keys)

  3. Create or copy your API key

  4. Update your mcp.json file:

    • Location: ~/.cursor/mcp.json (or C:\Users\<YourUsername>\.cursor\mcp.json on Windows)

    • Replace test_key_orcho_12345 with your actual API key

Test API Key

For initial testing, you can use:

test_key_orcho_12345

Note: The test key has limited functionality and rate limits. Get your own API key from app.orcho.ai for production use.

Security Best Practices

  • āœ… Store API keys only in ~/.cursor/mcp.json (not in your project)

  • āœ… Never commit API keys to version control

  • āœ… Rotate keys immediately if accidentally exposed


Usage

Manual Assessment

In Cursor chat, type:

@orcho assess_risk: Your prompt here

Enable automatic risk assessment for all prompts by adding a Cursor rule to your project.

Copy the rule file to your project:

# Create .cursor/rules directory
mkdir -p .cursor/rules

# Copy the rule file
cp node_modules/@orcho_risk/mcp-server/.cursor/rules/orcho-risk-assessment.mdc .cursor/rules/

Or manually copy from:

node_modules/@orcho_risk/mcp-server/.cursor/rules/orcho-risk-assessment.mdc

Option 2: Legacy .cursorrules File

Copy the example rules file:

cp node_modules/@orcho_risk/mcp-server/.cursorrules.example .cursorrules

Note: Project Rules (Option 1) are the modern approach and support more features.


How It Works

Context-Aware Assessment

Orcho automatically gathers context when available:

  • Current File: Detects the file open in your editor

  • Other Files: Analyzes which files will be modified by the prompt

  • Dependency Graph: Optional project dependency information

  • Blast Radius: Calculates impact scope of changes

Example

User: "Delete all user data from the database"
→ Cursor calls: @orcho assess_risk with context
→ Risk: HIGH (score: 95)
→ Cursor warns: "āš ļø HIGH RISK: This could cause data loss. Proceed?"

Tool Parameters

  • task (required): The prompt to assess

  • current_file (recommended): Path to currently open file

  • other_files (recommended): Array of files that will be modified

  • dependency_graph (optional): Project dependency graph

  • weights (optional): Custom risk calculation weights

  • aiignore_file (optional): Path to .aiignore file


Troubleshooting

MCP Server Not Loading

  1. Check mcp.json location:

    • Mac/Linux: ~/.cursor/mcp.json

    • Windows: C:\Users\<YourUsername>\.cursor\mcp.json

  2. Verify Node.js is installed:

    node --version  # Requires v18+
  3. Check Cursor Developer Tools:

    • Help → Toggle Developer Tools

    • Look for MCP-related errors in Console

API Errors

  • Invalid API Key: Verify the key is correct in mcp.json

  • Rate Limits: Check your account quota at app.orcho.ai

  • No API Key: The server will use the test key by default (limited functionality)

Still Having Issues?

  • Check that Cursor is fully restarted (quit and reopen)

  • Verify your API key is valid at app.orcho.ai

  • Ensure you have internet connectivity


License

MIT

Support

For issues and questions:

  • GitHub Issues: [Your Repo URL]

  • Orcho Support: [Support URL]

Available Tools

1 tool
assess_riskA

Assess the risk level of your coding prompt using Orcho risk analysis API. CRITICAL: You (Cursor AI) have access to the editor state - ALWAYS include context when available: 1) Pass the currently open/active file path as current_file (you can see this in the editor tabs), 2) Analyze the user prompt to determine which files will be modified and pass them as other_files array. Without context, only basic risk assessment is available. With context, you get blast radius and complexity analysis.

ParametersJSON Schema
NameRequiredDescriptionDefault
taskYesThe coding task or prompt you want to assess for risk.
current_fileNoSTRONGLY RECOMMENDED: Path to the currently open/active file in the Cursor editor (e.g., "src/main.js" or "mcp-server.js"). You (Cursor AI) can see which file is open in the editor tabs - always pass this if available. This enables context-aware assessment with blast radius and complexity analysis. If no file is open or unknown, omit this parameter.
other_filesNoSTRONGLY RECOMMENDED: Array of file paths that will be touched/modified by this prompt. Analyze the user prompt to determine which files will be affected (e.g., if prompt says "update login.js and auth.js", include ["login.js", "auth.js"]). If no other files will be touched, pass an empty array []. This enables accurate blast radius calculation. Always try to include this based on prompt analysis.
dependency_graphNoOptional JSON dependency graph of the project. Can be generated from package.json, requirements.txt, etc.
weightsNoOptional custom weights for risk calculation factors.
aiignore_fileNoOptional path to .aiignore file for excluding files from analysis.

TDQS

A4.3/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description carries the full burden. It effectively discloses key behavioral traits: the tool's dependency on context for enhanced analysis ('Without context, only basic risk assessment is available'), the importance of editor state access, and the specific outputs enabled by context ('blast radius and complexity analysis'). It doesn't mention rate limits or authentication needs, but covers the core operational behavior well.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is appropriately sized and front-loaded: the first sentence states the purpose, followed by critical usage instructions. Every sentence earns its place by providing essential guidance. It could be slightly more concise by integrating some details, but overall it's well-structured and efficient.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's complexity (6 parameters, nested objects) and no annotations or output schema, the description does a good job of explaining how to use it effectively. It covers the importance of context, parameter usage, and the enhanced analysis available. It doesn't describe the return format, but with no output schema, this is a minor gap rather than a critical omission.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the schema already documents all parameters thoroughly. The description adds some value by emphasizing the importance of 'current_file' and 'other_files' for context-aware assessment, but doesn't provide additional semantic details beyond what's in the schema. Baseline 3 is appropriate when the schema does the heavy lifting.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool's purpose: 'Assess the risk level of your coding prompt using Orcho risk analysis API.' It specifies the verb ('assess'), resource ('risk level of your coding prompt'), and method ('using Orcho risk analysis API'). With no sibling tools, this level of specificity is excellent.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides explicit usage guidance: 'CRITICAL: You (Cursor AI) have access to the editor state - ALWAYS include context when available.' It details when to use specific parameters (current_file when available, other_files based on prompt analysis) and explains the benefits of context ('With context, you get blast radius and complexity analysis').

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.

  1. 1 tool update
    • First observedassess_risk

TDQS

A4.1/5.0
Disambiguation5/5

With only one tool, there is no possibility of confusion or overlap between tools. The tool 'assess_risk' has a single, clearly defined purpose: to evaluate the risk level of coding prompts using the Orcho API. This eliminates any ambiguity in tool selection.

Naming Consistency5/5

The naming follows a consistent verb_noun pattern with 'assess_risk', which is clear and descriptive. Since there is only one tool, there is no inconsistency or mixing of conventions to evaluate, making it perfectly consistent by default.

Tool Count2/5

A single tool is generally too few for most server purposes, as it limits functionality and can feel thin. For a risk analysis server, one tool might suffice if it covers all necessary operations, but typical MCP servers benefit from multiple tools to handle different aspects of a domain, making this count borderline inadequate.

Completeness3/5

The tool 'assess_risk' provides a core risk assessment function, but there are notable gaps in coverage. For a risk analysis domain, one might expect additional tools for managing risk profiles, retrieving historical assessments, or configuring risk thresholds. However, the single tool does address the primary need, leaving room for expansion.

Maintenance

ActivityInactive
ResponsivenessNo issues

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Connectors

Related MCP Servers

  • A
    license
    Not graded
    quality
    D
    maintenance
    Automatically detects security vulnerabilities in AI-generated code, scanning for hardcoded secrets, injection flaws, XSS, weak cryptography, authentication issues, path traversal, and vulnerable dependencies across JavaScript, Python, Java, and Go.
    10
    2
    MIT
  • F
    license
    A
    quality
    D
    maintenance
    Enables deep security auditing of web applications directly from AI IDEs including Cursor and Claude Code. Scans URLs for vulnerabilities, returns security scores with SHIP/BLOCK verdicts, and provides specific fix prompts for remediation.
    3
    -
  • A
    license
    Not graded
    quality
    B
    maintenance
    Provides AI coding assistants with real-time security scanning superpowers, including SAST, secrets detection, dependency CVE scanning, and web vulnerability assessment.
    27
    Apache 2.0
  • F
    license
    Not graded
    quality
    D
    maintenance
    Securely feeds summarized expert security rules into your coding assistance Claude Code, Cursor, etc — zero config, no API key.
    -

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/guardd/mcp_server'

If you have feedback or need assistance with the MCP directory API, please join our Discord server