Detects exposed Discord tokens and credentials in code and files
Detects exposed GitHub tokens and credentials in code and files
Detects exposed GitLab tokens and credentials in code and files
Detects exposed Heroku API keys and credentials in code and files
Detects exposed MongoDB connection strings and credentials in code and files
Detects exposed OpenAI API keys and credentials in code and files
Detects exposed PostgreSQL connection strings and credentials in code and files
Detects exposed Redis connection strings and credentials in code and files
Detects exposed Slack tokens and credentials in code and files
Detects exposed Square API keys and credentials in code and files
Detects exposed Stripe API keys and credentials in code and files
Detects exposed Vercel tokens and credentials in code and files
credential-free
MCP server that scans code for exposed secrets. Detects API keys, tokens, and credentials in 69 patterns across cloud, AI, payment, and auth services.
Quick Setup ( 2 minutes )
Install System-wide
⚠️ Requires Python 3.10+
Alternative: Requirements File
Add to Your IDE
Windsurf / Kiro
Edit mcp_config.json:
Find your Python path:
Claude Desktop
Edit claude_desktop_config.json:
Test It
MCP Server Test
CLI Test
Usage Examples
MCP Server (AI Assistant)
Ask your AI:
"scan this file for secrets"
"check my project for exposed api keys"
"what patterns do you detect?"
"find any hardcoded credentials"
"scan my project but exclude .txt files"
CLI (Standalone)
tools
MCP Tool | CLI Command | description |
|
| scan a file for secrets |
|
| scan a folder recursively (supports exclude_patterns) |
|
| scan text directly |
|
| list detection patterns |
what it detects
69 patterns across:
cloud: aws, gcp, azure, vercel, heroku
ai: openai, anthropic, huggingface, groq
payment: stripe, square
auth: github, gitlab, slack, discord
database: postgresql, mongodb, redis
example output
docker
built for aws global vibe hackathon 2025