MaxKB
π₯ MaxKB is an open-source platform for building enterprise-grade agents. 强倧ζη¨ηεΌζΊδΌδΈηΊ§ζΊθ½δ½εΉ³ε°γ
0Tools
4Findings
20.6kStars
βDownloads
Mar 19, 2026Last Scanned
Findings4
1critical
2high
0medium
1low
0informational
criticalK5Auto-Approve / Bypass Confirmation PatternMCP06-excessive-permissionsAML.T0054
Pattern "\-\-(yes|force|no-confirm|no-prompt|assume-yes)\b" matched in source_code: "--force" (at position 1253)
Never auto-approve or bypass human confirmation for operations with side effects. Implement explicit confirmation gates that cannot be programmatically bypassed. If batch/CI mode is needed, require an explicit opt-in flag with audit logging. Required by EU AI Act Art. 14 (human oversight) and OWASP ASI09.
highC8No Auth on Network InterfaceMCP07-insecure-config
Pattern "0\.0\.0\.0" matched in source_code: "0.0.0.0" (at position 1978)
Add authentication middleware before exposing the server on a network interface. Use API keys, JWT tokens, or OAuth for authentication. Never expose MCP servers on 0.0.0.0 without auth.
highP10Host Network Mode and Missing Egress ControlsMCP07-insecure-configT1557
Pattern "0\.0\.0\.0:\d{2,5}(?!.*(?:NetworkPolicy|firewall|iptables|security[_\-]?group))" matched in source_code: "0.0.0.0:8080" (at position 1978)
Never use host network mode for MCP server containers β it bypasses all network isolation, exposing host services and enabling ARP/mDNS attacks. Use bridge networking with explicit port mapping. In Kubernetes, set hostNetwork: false and apply NetworkPolicy resources to restrict both ingress and egress traffic. Implement egress controls to prevent MCP servers from scanning internal networks or reaching unintended external services.
lowF4MCP Spec Non-ComplianceMCP07-insecure-config
Server fails MCP spec compliance checks: required:server_name; required:server_version; required:protocol_version; recommended:tool_descriptions; recommended:parameter_descriptions
Follow the MCP specification for server metadata. Include server name, version, and protocol version. Provide descriptions for all tools and parameters.
Tools
No tools exposed by this server.
Security Category Deep Dive
Sub-Category Tree Β· Remediation Roadmap Β· Attack Stories Β· Compliance Overlay Β· ATLAS Techniques Β· Maturity Model
Prompt Injection
Prompt & context manipulation attacks
69
Maturity
14
Rules
5
Sub-Categories
1
Gaps
64%
Implemented
56
Tests
1
Stories
100%3 rules
Injection via tool descriptions and parameter fields
GAP-001Prompt Injection Coverage GapMissing detection coverage for emerging prompt injection attack variants not addressed by current rules
100%4 rules
Hidden instructions via external content and tool responses
100%2 rules
Context window saturation and prior-approval exploitation
100%3 rules
Payload hiding via invisible chars, base64, schema fields
100%2 rules
Injection via prompt templates and runtime tool output