Guardianshield
Adds a security scanning layer to Claude with 27 tools covering code vulnerabilities, dependency CVEs, PII detection, prompt injection defense, and secret scanning. The code scanner catches SQL injection, XSS, and command injection with CWE mappings and auto-fix suggestions, while the DeepEngine performs cross-line taint tracking in Python. Dependencies get checked against OSV.dev for PyPI, npm, Go, and Packagist packages. You get SARIF export for GitHub Code Scanning, baseline comparison for delta scanning, quality gates with severity thresholds, and SQLite-backed audit logs. Comes with five safety profiles for different compliance contexts. Reach for this when you need to scan code or validate AI inputs and outputs for security issues directly in your Claude workflow.
- GitHub stars
- 3
- tools
- 27
- transport
- STDIO
- env vars
- 3
- last updated
- Mar 3, 2026
Tools
27 toolsPublic tool metadata for what this MCP can expose to an agent.
scan_codeScan source code for vulnerabilities and hardcoded secrets
Scan source code for vulnerabilities and hardcoded secrets
No parameter schema in public metadata yet.
scan_fileScan a single file (auto-detects language from extension)
Scan a single file (auto-detects language from extension)
No parameter schema in public metadata yet.
scan_directoryRecursively scan a directory with filtering and progress streaming
Recursively scan a directory with filtering and progress streaming
No parameter schema in public metadata yet.
scan_inputCheck user/agent input for prompt injection attempts
Check user/agent input for prompt injection attempts
No parameter schema in public metadata yet.
scan_outputCheck AI output for PII leaks and content violations
Check AI output for PII leaks and content violations
No parameter schema in public metadata yet.
check_secretsDetect hardcoded secrets and credentials
Detect hardcoded secrets and credentials
No parameter schema in public metadata yet.
scan_filesScan multiple files in one call
Scan multiple files in one call
No parameter schema in public metadata yet.
scan_diffParse unified diff and scan only added lines
Parse unified diff and scan only added lines
No parameter schema in public metadata yet.
check_dependenciesCheck packages for known CVEs via OSV.dev (PyPI, npm, Go, Packagist)
Check packages for known CVEs via OSV.dev (PyPI, npm, Go, Packagist)
No parameter schema in public metadata yet.
sync_vulnerabilitiesSync the local OSV vulnerability database
Sync the local OSV vulnerability database
No parameter schema in public metadata yet.
parse_manifestParse any supported manifest file (11 formats) into dependency objects
Parse any supported manifest file (11 formats) into dependency objects
No parameter schema in public metadata yet.
scan_dependenciesScan a directory for manifest files and check all deps for vulnerabilities
Scan a directory for manifest files and check all deps for vulnerabilities
No parameter schema in public metadata yet.
mark_false_positiveMark a finding as false positive (flags future matches)
Mark a finding as false positive (flags future matches)
No parameter schema in public metadata yet.
list_false_positivesList active false positive records with optional filter
List active false positive records with optional filter
No parameter schema in public metadata yet.
unmark_false_positiveRemove a false positive record by fingerprint
Remove a false positive record by fingerprint
No parameter schema in public metadata yet.
list_enginesList available analysis engines with capabilities
List available analysis engines with capabilities
No parameter schema in public metadata yet.
set_engineSet active analysis engines for code scanning
Set active analysis engines for code scanning
No parameter schema in public metadata yet.
export_sarifExport findings as SARIF 2.1.0 JSON for GitHub Code Scanning and CI
Export findings as SARIF 2.1.0 JSON for GitHub Code Scanning and CI
No parameter schema in public metadata yet.
save_baselineSave current findings as a baseline for delta scanning
Save current findings as a baseline for delta scanning
No parameter schema in public metadata yet.
scan_with_baselineScan code and report only new findings vs. baseline
Scan code and report only new findings vs. baseline
No parameter schema in public metadata yet.
check_quality_gateEvaluate findings against severity thresholds (pass/fail/warn)
Evaluate findings against severity thresholds (pass/fail/warn)
No parameter schema in public metadata yet.
get_profileGet current safety profile configuration
Get current safety profile configuration
No parameter schema in public metadata yet.
set_profileSwitch safety profile (general, education, healthcare, finance, children)
Switch safety profile (general, education, healthcare, finance, children)
No parameter schema in public metadata yet.
test_patternTest a regex pattern against sample code for custom pattern development
Test a regex pattern against sample code for custom pattern development
No parameter schema in public metadata yet.
audit_logQuery the security audit log
Query the security audit log
No parameter schema in public metadata yet.
get_findingsRetrieve past findings with filters
Retrieve past findings with filters
No parameter schema in public metadata yet.
shield_statusGet health, configuration, and OSV cache statistics
Get health, configuration, and OSV cache statistics
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageConfiguration
GUARDIANSHIELD_PROFILEDefault safety profile (general, education, healthcare, finance, children)
GUARDIANSHIELD_AUDIT_PATHPath to the SQLite audit database
GUARDIANSHIELD_DEBUGSet to 1 for debug logging
Readme
GuardianShield
Universal AI security layer — an open-source MCP server for code scanning, PII detection, prompt injection defense, secret detection, dependency auditing, and audit logging.
Zero dependencies · 27 MCP tools · 5 safety profiles · 108+ detection patterns
Features
Details
- Repository
- github.com/sparkvibe-io/guardianshield
- Categories
- AI & LLM Tools, Security & Pentesting
- Package
guardianshield- Transport
- STDIO
- Registry
- MCP registry, active
Related AI & LLM Tools MCP servers
View all
Mcpwireboard/mcpOfficial MCP server for WireBoard. Lets LLM agents query your analytics in conversation.0stars
Draw2agentzero-abd/draw2agentDraw on your website, let your AI agent see it. Visual context for code generation via MCP.3stars
Mcpcueapi/cueapi-mcpSchedule agent work and report write-once outcomes via CueAPI from any MCP host.2stars
Afterpathsburnssa/afterpathsSession memory for AI coding agents. Search past sessions, extract rules, and track what works.2stars
Wisdomgraphcklam12345/wisdomgraphNeo4j-native DIKW memory for Claude Code — knowledge that compounds across sessions2stars
LittleSis API Servercode-rabi/littlesis-mcpProvide access to the LittleSis API to track corporate power and accountability. Enable querying and exploring relationships and entities related to corporate i2stars