Skip to content
VLN Logo VLN Documentation

Skill-MCP FAQ & Troubleshooting

Skill-MCP FAQ & Troubleshooting

Common questions and solutions for Skill-MCP.

General Questions

What is Skill-MCP?

Skill-MCP is VLN’s implementation of the Model Context Protocol (MCP) providing 50+ tools for smart contract analysis, gaming analytics, blockchain operations, and data integration.

How many tools are available?

Currently 50+ tools across 4 categories:

  • Security & Analysis (15 tools)
  • Blockchain Operations (12 tools)
  • Gaming Analytics (10 tools)
  • Data & Integration (13 tools)

What languages are supported?

  • JavaScript/TypeScript (native)
  • Python (full SDK)
  • Go (Go SDK)
  • REST API (all languages)

What’s the pricing model?

  • Free tier: 100 requests/minute, 10 concurrent
  • Pro: 1000 requests/minute, 50 concurrent
  • Enterprise: Unlimited with custom SLA

Contact sales@vln.gg for details.

Setup & Installation

Q: How do I install Skill-MCP?

JavaScript:

Terminal window
npm install @vln/skill-mcp

Python:

Terminal window
pip install vln-skill-mcp

Go:

Terminal window
go get github.com/Fused-Gaming/skill-mcp-go

Q: I get “No API key provided” error

A: Set environment variable:

Terminal window
export SKILL_API_KEY="sk_live_xxxxx"

Or pass directly to client:

const client = new SkillClient({
  apiKey: 'sk_live_xxxxx'
});

Q: Where do I get an API key?

A:

  1. Visit https://skill-mcp.vln.gg/dashboard
  2. Click “API Keys”
  3. Generate new key
  4. Copy and store securely

Q: Can I use the same key for dev and production?

A: Not recommended. Generate separate keys:

  • sk_test_... for development
  • sk_live_... for production

Q: How do I rotate API keys?

A: In dashboard:

  1. Go to API Keys
  2. Click “Rotate” on existing key
  3. Old key becomes invalid immediately
  4. Update all applications

Tool Usage

Q: How do I list available tools?

A:

JavaScript:

const tools = await client.tools.list();
tools.forEach(tool => console.log(tool.name));

Python:

tools = client.tools.list()
for tool in tools:
    print(tool['name'])

REST API:

Terminal window
curl https://skill-mcp.api.vln.gg/api/tools \
  -H "Authorization: Bearer sk_live_..."

Q: What does the “analyze-contract” tool do?

A: Scans a smart contract for security vulnerabilities:

const analysis = await client.tools.execute('analyze-contract', {
  contractAddress: '0x...',
  chainId: 1
});


// Returns:
// - vulnerabilities array
// - riskScore (0-100)
// - recommendations

Q: How long do tool executions take?

A: Typical execution times:

  • analyze-contract: 2-5 seconds
  • assess-wallet-risk: 1-2 seconds
  • fetch-market-data: 500ms-1s
  • Check tool documentation for specifics

Q: Can I execute multiple tools concurrently?

A: Yes, up to your concurrency limit:

JavaScript:

const results = await Promise.all([
  client.tools.execute('analyze-contract', params1),
  client.tools.execute('assess-wallet-risk', params2),
  client.tools.execute('fetch-market-data', params3)
]);

Q: What if a tool returns an error?

A: Handle errors gracefully:

try {
  const result = await client.tools.execute('analyze-contract', {
    contractAddress: address
  });
} catch (error) {
  if (error.code === 'INVALID_PARAMS') {
    console.error('Invalid parameters:', error.details);
  } else if (error.code === 'RATE_LIMITED') {
    // Wait and retry
  } else {
    console.error('Unexpected error:', error);
  }
}

Rate Limiting & Quotas

Q: How do I check my quota?

A:

const quotas = await client.quotas.get();
console.log(`Remaining: ${quotas.remaining} / ${quotas.limit}`);

Q: What happens if I exceed my quota?

A: You’ll get a 429 error:

{
  "error": {
    "code": "RATE_LIMITED",
    "message": "Quota exceeded",
    "retryAfter": 60
  }
}

Q: Can I request a higher quota?

A: Yes! Email sales@vln.gg with:

  • Current usage
  • Projected needs
  • Use case description

Q: Are there per-minute and per-day limits?

A: Yes:

  • Free: 100 req/min, 10,000 req/day
  • Pro: 1000 req/min, 1M req/day
  • Enterprise: Custom limits

Q: How do I avoid rate limiting?

A:

  1. Batch operations when possible
  2. Cache results (10+ minute TTL)
  3. Implement exponential backoff
  4. Monitor quota usage
  5. Upgrade tier if needed

API & Integration

Q: Do you have a REST API?

A: Yes! Full REST API:

Terminal window
curl -X POST https://skill-mcp.api.vln.gg/api/tools/execute \
  -H "Authorization: Bearer sk_live_..." \
  -H "Content-Type: application/json" \
  -d '{
    "toolName": "analyze-contract",
    "parameters": {
      "contractAddress": "0x..."
    }
  }'

Q: Can I use Skill-MCP with my LLM/AI?

A: Yes! Skill-MCP is designed for LLM integration. Tools work seamlessly with:

  • Claude API
  • OpenAI API
  • Custom AI systems

Q: How do I integrate with FastAPI?

A:

from fastapi import FastAPI
from skill_mcp import SkillClient


app = FastAPI()
skill = SkillClient(api_key='sk_live_...')


@app.post("/analyze")
async def analyze(address: str):
    return await skill.tools.execute('analyze-contract', {
        'contractAddress': address
    })

Q: Can I use webhooks?

A: Yes, for event-driven integration:

Terminal window
curl -X POST https://skill-mcp.api.vln.gg/api/webhooks \
  -H "Authorization: Bearer sk_live_..." \
  -d '{
    "url": "https://your-app.com/webhooks/skill",
    "events": ["execution.success", "execution.failed"]
  }'

Troubleshooting

Q: I get “Tool not found” error

A:

  1. Check tool name is correct
  2. List available tools to verify
  3. Some tools may require additional setup

Q: Execution times out

A:

  1. Check tool documentation for timeout
  2. Try with simpler parameters
  3. Contact support if persistent

Q: I keep getting “Invalid address” error

A:

  • Verify address format (must be 0x + 40 hex chars)
  • Check for extra spaces
  • Ensure address is checksummed correctly
// Correct format
'0x1234567890123456789012345678901234567890'


// Wrong formats
'0x12345678901234567890123456789012345678'  // Too short
'12345678901234567890123456789012345678'    // Missing 0x

Q: Results are cached, how do I force refresh?

A:

  • Cache TTL is 10 minutes
  • Results are automatically refreshed after TTL
  • For immediate refresh, include skipCache: true:
const result = await client.tools.execute('analyze-contract', {
  contractAddress: address,
  skipCache: true  // Force fresh analysis
});

Q: Tool execution is slow

A:

  1. Check typical execution time in docs
  2. Verify network connectivity
  3. Check current API load (dashboard)
  4. Try again in a few minutes
  5. Contact support if consistently slow

Q: How do I debug tool execution?

A: Enable debug logging:

const client = new SkillClient({
  apiKey: process.env.SKILL_API_KEY,
  debug: true  // Enable verbose logging
});

Custom Tools

Q: Can I create custom tools?

A: Yes! Custom tools are available for Pro/Enterprise:

  1. Request custom tool development
  2. Provide tool specification
  3. VLN develops and deploys
  4. Access via standard API

Contact sales@vln.gg for details.

Q: How do I use a custom tool?

A: Same as built-in tools:

const result = await client.tools.execute('custom-my-tool', params);

Data & Privacy

Q: What data do you store?

A:

  • Execution history (30 days)
  • Quotas and usage (forever)
  • API keys (hashed)
  • Logs for debugging (30 days)

Q: Is my data encrypted?

A: Yes:

  • In transit: TLS 1.3
  • At rest: AES-256-GCM
  • No plaintext logs

Q: Can I delete my execution history?

A: Contact privacy@vln.gg with request.

Q: Is Skill-MCP GDPR compliant?

A: Yes, fully GDPR compliant with DPA.

Performance & Limits

Q: What’s the maximum request size?

A: 1MB per request

Q: What’s the maximum response size?

A: 10MB per response

Q: Can I process large datasets?

A: For bulk processing:

  1. Batch requests (10-100 per batch)
  2. Implement pagination
  3. Use webhook callbacks
  4. Contact sales for bulk pricing

Getting Help

Need assistance?