secscan-mcp
The secscan-mcp server provides comprehensive security scanning tools for AI coding assistants, enabling detection of secrets, code vulnerabilities, dependency issues, and IaC misconfigurations.
Secret Detection (
scan_secrets): Scans a directory for hardcoded credentials, with an option to include git commit history.SAST Analysis (
scan_code): Runs static analysis tools (semgrep, bandit) to find code-level security vulnerabilities.Dependency Scanning (
scan_dependencies): Checks lockfiles for known vulnerable packages using osv-scanner.IaC Scanning (
scan_iac): Detects misconfigurations in Infrastructure-as-Code files via checkov.Unified Scanning (
scan_all): Runs all available scanners in one call and produces a consolidated, normalized report.Scanner Discovery (
list_available_scanners): Reports which third-party scanning CLIs are installed and available.Remediation Guidance (
explain_finding): Provides detailed explanations and remediation hints for a specificrule_idfrom scan results.
All scan tools accept a path parameter and an optional severity_threshold (critical, high, medium, low, info) to filter results. The server operates gracefully, skipping optional scanners if their CLIs are not installed, and works with any MCP-compatible AI assistant (e.g., Cursor, VS Code, Claude Desktop).
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@secscan-mcprun a full security scan on this project with high severity threshold"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
secscan-mcp
A portable MCP server for security scanning — works with any AI coding assistant that supports the Model Context Protocol: Cursor, VS Code, Claude Desktop, Windsurf, Zed, Continue, and more.
Scan codebases for hardcoded secrets, SAST issues, vulnerable dependencies, and IaC misconfigurations — one install, one normalized report format.
The built-in custom scanner works with no extra tools. Install optional CLIs for broader coverage (below).
Quick start
Requires Python 3.11+. If pip install secscan-mcp says "No matching distribution found", your default python3 is likely too old — use python3.11 -m pip install secscan-mcp or install Python 3.11+ first.
1. Install from PyPI:
pip install secscan-mcp
# or explicitly:
python3.11 -m pip install secscan-mcpOr run without installing (requires uv):
uvx secscan-mcpFor MCP config with uvx, use "command": "uvx" and "args": ["secscan-mcp"] — see setup guide.
git clone https://github.com/openjkai/secscan_mcp.git
cd secscan_mcp && pip install .2. Add to your IDE — pick your client:
IDE / client | Config file | Guide |
Cursor |
| |
VS Code |
| |
Claude Desktop | OS-specific (see guide) | |
Claude Code |
| |
Windsurf |
| |
Others | — |
Minimal config (works in Cursor, Claude Desktop, Windsurf):
{
"mcpServers": {
"secscan": {
"command": "uvx",
"args": ["secscan-mcp"]
}
}
}If you installed with pip install secscan-mcp, you can use "command": "secscan-mcp" instead.
3. Verify — ask your agent: "Call list_available_scanners and scan_secrets on this project."
Related MCP server: stepsecurity-mcp
MCP tools
Tool | Purpose |
| Which engines are installed on this machine |
| Hardcoded credentials and secrets (optionally scan git commit history) |
| SAST (semgrep, bandit) |
| Vulnerable packages (osv-scanner) |
| IaC misconfigurations (checkov) |
| All available scanners, one unified report |
| Remediation hints for a |
Most scan tools accept path (directory to scan) and optional severity_threshold (critical, high, medium, low, info).
scan_secrets also accepts include_git_history (boolean). When true, scans past git commits for secrets removed from the working tree but still present in history — no extra tools required beyond git. scan_all accepts include_git_history too.
Suppressing false positives
Silence known-good findings without changing scanner behavior. Both mechanisms apply to every engine, and each report includes a suppressed count so nothing is hidden silently.
.secscanignore at the project root — gitignore-style path globs:
# ignore vendored code and test fixtures
vendor/
tests/fixtures/
*.min.jsInline # nosecscan on the offending source line — suppress all rules there, or scope to specific rule IDs:
API_TOKEN = get_token() # real code, no marker
LEGACY_KEY = "AKIA..." # nosecscan
DEMO_JWT = "eyJ..." # nosecscan: hardcoded-jwtOptional scanners
Install any of these to extend coverage. Missing CLIs are skipped — the server still runs.
Engine | Category | Install (example) |
gitleaks | secrets |
|
semgrep | SAST |
|
bandit | SAST (Python) |
|
osv-scanner | dependencies |
|
checkov | IaC |
|
After installing, run list_available_scanners again to confirm.
Example prompts
"Call
list_available_scannersand tell me what's installed.""Run
scan_secretswith include_git_history on this repo — check if any secrets were ever committed.""Run
scan_allwith severity_threshold high and summarize the findings.""Explain the rule
internal-api-key.""Add a
.secscanignorefor thetests/fixturesdirectory and re-run the scan."
Configuration
Environment variables (optional):
Variable | Default | Description |
|
| Per-engine scan timeout |
|
| Max findings per report |
|
| Max commits scanned in git history mode |
Pass via MCP config env block — see setup guide.
Development
make install-dev # editable install + dev tools
make check # lint + typecheck + testSee docs/CONTRIBUTING.md and PLAN.md.
License
MIT
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- AlicenseAqualityDmaintenanceA security-focused server that integrates with Cursor IDE to provide real-time vulnerability detection, exploit generation, and security insights during software development.Last updated72MIT

stepsecurity-mcpofficial
AlicenseAqualityDmaintenanceMCP server for the StepSecurity platform that enables investigating supply-chain and CI/CD security issues through natural language.Last updated3063Apache 2.0- Alicense-qualityCmaintenanceMCP server for ai-scanner that enables AI agents to scan codebases for LLM usage, AI frameworks, and exposed secrets.Last updated531MIT
- Flicense-qualityDmaintenanceMCP server that scans project dependencies for security vulnerabilities (CVEs) and provides fix instructions directly in VS Code via Copilot.Last updated3
Related MCP Connectors
Security scanner for MCP servers. Detect vulnerabilities, prompt injection, and tool poisoning.
Zero-config MCP security scanner for AI-generated apps. 25K+ vulnerability patterns.
Scan any public GitHub MCP-server repo for security issues. 37 MCP-specific L1 rules, 8 languages.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
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/openjkai/secscan_mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server