About TridentChain Security
TridentChain Security is a local-first scanner for project dependencies, developer tools, and IDE extensions. It suits developers and agents that need vulnerability checks without uploading project data, using advisories from several sources with KEV/EPSS prioritization.
Key features
- 01Scans project dependencies, system tools, and IDE extensions
- 02Combines OSV, NVD, GHSA, and Sonatype data
- 03Prioritizes findings with KEV/EPSS
- 04Includes quick, full, and offline profiles
- 05CLI, Python API, desktop app, and agent integrations
Use cases
- 01Check a project for known dependency vulnerabilities
- 02Scan IDE extensions and developer tools
- 03Run local-only scans without network access
- 04Validate a code change through an agent
MCP support
The project provides an MCP server for agent integrations. MCP tools expose scanner capabilities, including validation after a patch.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
3Saved tool list · updated 2w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| output_dir | any | No | |
| run_profile | string | No | |
| max_findings | integer | No | |
| project_path | string | Yes |
Details
From the project READMETridentChain Security
Local-first vulnerability scanner for project dependencies, developer tools, and IDE extensions.
Uses multi-source intelligence (OSV, NVD, GHSA, Sonatype) with KEV/EPSS prioritization.
No API key required for default usage.
Public repo: https://github.com/DevInder1/supply-chain-scanner-public
Install (plug and play)
pip3 install tridentchain-security
npm install -g @tridentchain/security-cli
tridentchain-security --help
Agents & MCP (Claude, Cursor, VS Code, Windsurf, Zed):
Pick whichever install path fits:
# Option A — pip (needs Python 3.10+)
pip3 install -U "tridentchain-security>=0.1.4" "tridentchain-mcp>=0.1.4"
# Option B — uvx (no manual Python install; uv handles it)
uvx tridentchain-mcp
# Option C — Docker (no Python needed, fully sandboxed)
docker pull ghcr.io/devinder1/tridentchain-mcp:latest
# Then in your MCP config, replace `python3 -m tridentchain_mcp` with:
# command: docker
# args: ["run", "--rm", "-i", "-v", "$PWD:/workspace", "ghcr.io/devinder1/tridentchain-mcp:latest"]
What you can do: docs/CAPABILITIES.md
Full guide: docs/INSTALL_AND_USE.md
Cross-platform (macOS / Linux / Windows): docs/CROSS_PLATFORM.md
(PyPI: tridentchain-security · npm: @tridentchain/security-cli)
tridentchain-security --scan all --project-path . --output-dir scanner-output
Use in your own Python app
from scanner import run_scan
summary = run_scan(
project_path=".",
scan="all",
run_profile="full", # no API key required
output_dir="scanner-output",
)
print(summary["summary"])
Scan profiles
| Profile | Description |
|---|---|
full (default) | Project + system + extensions. OSV + NVD without keys. |
quick | Faster project-focused scan. |
offline | Local advisory DB only, no network. |
| Power-user | Add GITHUB_TOKEN, NVD_API_KEY, optional SONATYPE_TOKEN for best coverage. |
Desktop app (individual application)
No repo clone required if the pip package is installed:
pip3 install tridentchain-security
cd apps/desktop && npm install && npm run start
See apps/desktop/README.md and docs/DISTRIBUTION_VERIFICATION.md.
AI / automation (Claude, OpenAI, Cursor, VS Code, Windsurf, …)
One install, every agent: pip install "tridentchain-security>=0.1.2" tridentchain-mcp
| Guide | Description |
|---|---|
| Agent integrations | Claude · OpenAI · Cursor · VS Code · Windsurf · Zed · MCP · CLI |
| Capabilities | Everything you can do today |
| Architecture | MCP + unified tools design |
./scripts/setup-agent-mcp.sh cursor # prints setup for your agent
Phase 2 — Claude MCP: pip install tridentchain-mcp · Setup guide · Plugin
Phase 3 — OpenAI + Cursor: examples/openai/ · Cursor setup · .cursor/mcp.json.example
Phase 4 — VS Code (Anthropic MCP): Open repo → MCP ready · VS Code setup · ./scripts/vscode-mcp-install-link.sh · extension
Phase 5 — Validate: tridentchain-security --validate · MCP validate_after_patch · CAPABILITIES.md
Unified tool layer: from scanner.integrations import execute_tool, get_tool_definitions, to_openai_tools
Development
git clone https://github.com/DevInder1/supply-chain-scanner-public.git
cd supply-chain-scanner-public
python3 -m pip install -e .
tridentchain-security --help
python3 -m unittest scanner.tests.test_matcher_ranges -v
Install & use: docs/INSTALL_AND_USE.md
Cross-platform: docs/CROSS_PLATFORM.md
CLI contract: docs/cli-contract.md
Publishing: docs/PUBLISHING.md
Optional API keys (power users)
| Variable | Purpose |
|---|---|
NVD_API_KEY | Higher NVD rate limits |
GITHUB_TOKEN | GHSA advisories |
SONATYPE_TOKEN | Sonatype Guide advisories |
Set in .env or environment variables.
License
MIT — see LICENSE
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.