finktech-dev/MCP Verify

Validate, test, and secure Model Context Protocol (MCP) servers with comprehensive security scanning

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Stale
License
None
namedescriptionrequireddefault
serverServer command to validate (e.g., "node server.js" or "python server.py")yes
transportTransport type (http or stdio)nostdio
formatOutput format: json, sarif, htmlnosarif
output-dirOutput directory for reportsnomcp-reports
enable-fuzzingEnable intelligent fuzzing for chaos testingnofalse
enable-semanticEnable LLM-powered semantic analysis (requires ANTHROPIC_API_KEY)nofalse
llm-providerLLM provider for semantic analysis (e.g., "anthropic:claude-haiku-4-5-20251001", "ollama:llama3.2")noanthropic:claude-haiku-4-5-20251001
rulesSecurity rule blocks to execute (e.g., "OWASP,MCP,A")noOWASP,MCP,A,B,C
exclude-rulesRule blocks or IDs to exclude (e.g., "SEC-001,D")no
min-severityMinimum severity level to report (info, low, medium, high, critical)nomedium
fail-on-issuesFail the workflow if security issues are foundnotrue
baseline-pathPath to baseline report for regression detectionno
fail-on-degradationFail if security score drops compared to baselinenofalse
allowed-score-dropAllowed security score drop before failing (0-100)no5
namedescription
report-pathPath to the generated report file
security-scoreOverall security score (0-100)
critical-issuesNumber of critical security issues found
high-issuesNumber of high severity issues found