Security Suite

io.github.4hmetuyarPublished by GitHub account 4hmetuyarv0.2.3Updated Nov 10, 2026

Bundle of secret-scanner, dependency-auditor, ssl-inspector, and dns-intelligence

Overview

AI-generated overview

GuardBee Security Suite bundles secret, dependency, TLS, DNS and AI-agent security scanners into one local MCP server and CLI.

What you can do
  • Scan code and content for secrets, prompt injection, tool poisoning, PII and OWASP MCP Top 10 issues
  • Audit dependencies for CVEs, hallucinated packages, model files and MCP client configs
  • Inspect hosts over the network: TLS certificates, DNS and email records, exposed vector stores
  • Run jailbreak probes against an LLM endpoint you control
  • Call GuardBee platform scans, compliance checks and fix suggestions when an API key is set
  • Run the same checks from a CLI with SARIF output for CI pipelines
When to use it
Useful when you want an assistant to audit a codebase, MCP server or agent setup for security problems, or to check a domain's TLS, DNS and email configuration, without wiring up several separate scanners.
Requirements
Runs locally over stdio via npx from the npm package @guardbee/security-suite. No credentials are needed for the default recommended tool set; the platform group (vulnerability scans, compliance checks, security engineer fixes) requires the GUARDBEE_API_KEY environment variable. Network access is needed for the network and supply-chain checks.
Before you install
The platform group needs GUARDBEE_API_KEY and sends data to the GuardBee API. Network tools contact external hosts, and the LLM red-team tools send probes to an endpoint you specify. Telemetry is off by default but enabling GUARDBEE_TELEMETRY=1 sends tool names, success, duration and short parameter values such as file paths. Tool names changed from earlier versions, so existing agent prompts may need updating.

Installation

In SourceWeft

  1. Open Security Suite in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

@guardbee/security-suite

🇬🇧 English | 🇹🇷 Türkçe | 🇨🇳 中文 | 🇯🇵 日本語 | 🇰🇷 한국어 | 🇧🇷 Português | 🇪🇸 Español | 🇩🇪 Deutsch

[npm version] [npm downloads] [License: MIT]

Every GuardBee MCP security tool from one install. guardbee serve is one MCP server with the tools of the packages you pick; guardbee <package> … runs any package's CLI. 30 packages, 129 MCP tools.

Telemetry is off by default. With GUARDBEE_TELEMETRY=1 it sends the server, tool, success, duration and short parameter values such as a file path; longer strings and fields like content, token or password are replaced with a placeholder. See @guardbee/mcp-telemetry.

Quick start

json
{  "mcpServers": {    "guardbee": {      "command": "npx",      "args": ["-y", "@guardbee/security-suite", "serve"]    }  }}

That serves the recommended set (31 tools): a one-pass OWASP MCP Top 10 scan, secrets, prompt injection, toxic flows, MCP client configs and dependency CVEs. Pick others with --tools:

json
"args": ["-y", "@guardbee/security-suite", "serve", "--tools=recommended,network"]

--tools takes group names and package ids, comma-separated, or all (129 tools; some clients cap the number of tools, Cursor among them). GUARDBEE_SUITE_TOOLS does the same as an environment variable. Each tool's name starts with its package's prefix, so two packages' scan_text tools do not collide: secrets_scan_text, injection_scan_text. Names that already say it stay as they are (owasp_scan_path, inspect_ssl).

Groups

GroupWhat it covers
recommended (default)owasp-scan, secret-scanner, prompt-injection-scanner, toxic-flow-auditor, mcp-config-auditor, dependency-auditor
codeStatic checks of source code, configs and content on disk
supply-chainDependencies, hallucinated packages, model files, MCP client configs, tool-definition drift
networkChecks that reach out to a host: TLS, DNS, vector stores, a live LLM
platformTools that call the GuardBee platform (need GUARDBEE_API_KEY)
allEvery package

Packages

IdTool prefixChecksGroups
owasp-scanowasp_OWASP MCP Top 10 in one pass over a codebase or a live server, A–F graderecommended, code
secret-scannersecrets_API keys, tokens and passwords in files, git history and agent transcriptsrecommended, code
prompt-injection-scannerinjection_Prompt injection in documents, web pages and RAG contentrecommended, code
toxic-flow-auditortoxic_flow_Lethal trifecta: untrusted input, sensitive data and egress in one agentrecommended, code
mcp-config-auditormcp_config_MCP client configs (Cursor, Claude, Windsurf, VS Code) and shadow serversrecommended, supply-chain
dependency-auditordeps_Known CVEs in npm and PyPI dependencies (OSV)recommended, supply-chain
prompt-leak-scannerleak_Credentials and PII in outbound LLM prompts, before they leave your appcode
tool-poisoning-scannerpoisoning_Hidden instructions in MCP tool descriptions and schemascode
mcp-server-auditorserver_audit_Shell, eval, SSRF and secrets in an MCP server's sourcecode
oauth-auditoroauth_OAuth 2.1 mistakes in an MCP server's auth code: token passthrough, audience, PKCEcode
audit-gap-auditoraudit_gap_Tool calls without an audit trail (OWASP MCP08)code
context-oversharing-auditoroversharing_Context oversharing (MCP10): session dumps, cross-user context, unscoped recallscode
ai-code-scannerai_code_Insecure LLM integrations: exposed keys, unsafe model output, PII to third-party LLMscode
agent-graph-auditoragent_graph_Transitive excessive agency across LangGraph, CrewAI and AutoGen agentscode
a2a-auditora2a_Agent2Agent (A2A) implementation code: webhook SSRF, unauthenticated Agent Cardscode
elicitation-auditorelicitation_MCP elicitation that asks users for secrets or sends them to third partiescode
memory-poisoning-scannermemory_Untrusted input written into persistent agent memorycode
unbounded-consumption-auditorconsumption_Denial of wallet: missing token limits, timeouts and loop boundscode
threat-rulesthreat_rules_Agent Threat Rules (ATR) on text, files and agent eventscode
document-scannerdocument_PII and credentials in PDFs and images (OCR) before they reach an LLMcode
reverse-shieldreverse_How easily a shipped app can be reverse engineered (web, Electron, APK/AAB, IPA, Unity, native); fix diffs, CI baselinescode
slopsquat-scannerslopsquat_Hallucinated or brand-new package names (slopsquatting)supply-chain
model-scannermodel_Malicious pickle, joblib, NumPy, safetensors, Keras, ONNX, GGUF and TensorFlow model files, and Hugging Face repossupply-chain
rug-pull-detectorrug_pull_MCP tool definitions that change after you approved themsupply-chain
ssl-inspectorssl_TLS certificates, ciphers and HSTSnetwork
dns-intelligencedns_DNS records, SPF/DMARC/DKIM, subdomain takeovernetwork
vector-store-scannervector_store_Unauthenticated vector databases (Weaviate, Qdrant, Chroma, pgvector…)network
llm-redteamredteam_Jailbreak and extraction probes against an LLM endpoint you controlnetwork
vulnerability-scannervuln_GuardBee platform security scans and their findingsplatform
compliance-checkercompliance_KVKK, GDPR and CCPA compliance of a website via the GuardBee APIplatform
security-engineerengineer_Fix GuardBee AI Security Engineer findings in your repo and verify the fixes (GUARDBEE_API_KEY)platform

guardbee list prints this table. The proxy (mcp-security-proxy) and the database gateway (mcp-db-gateway) are not in the suite: they sit in front of other servers rather than answering tool calls.

CLI

bash
npm install -g @guardbee/security-suite
guardbee list                                         # packages and groupsguardbee owasp-scan scan . --fail-on=high --format=sarif > owasp.sarifguardbee secret-scanner scan . --history              # any package's CLI, same argumentsguardbee mcp-config-auditor discover

The shortcuts from earlier versions still work:

ShortcutRuns
guardbee secret-scan <path>secret-scanner scan
guardbee dep-audit <dir>dependency-auditor audit
guardbee ssl-inspect <domain>…ssl-inspector inspect
guardbee dns-check <domain>dns-intelligence check

GitHub Actions

yaml
- run: npx -y @guardbee/security-suite owasp-scan scan . --format=sarif --fail-on=high > owasp.sarif- run: npx -y @guardbee/security-suite dep-audit . --fail-on=high- uses: github/codeql-action/upload-sarif@v3  if: always()  with:    sarif_file: owasp.sarif

Upgrading from 0.2

guardbee serve used to expose five tools from four packages with plain names (scan_directory, scan_file, audit_dependencies, inspect_ssl, check_dns). It now serves the recommended group with prefixed names: the same checks are secrets_scan_directory, secrets_scan_file, deps_audit_directory, and, with --tools=recommended,network, inspect_ssl and enumerate_dns / dns_check_email_security. Agents that call tools by name need the new names.

License

MIT — GuardBee

Feedback

Questions, ideas, or a false positive? Start a thread in GitHub Discussions or open an issue. If GuardBee is useful to you, a ⭐ on the repository helps others find it.

Source: packages/security-suite/README.md at commit 462e3cb

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v0.2.3LatestSep 29, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts