GuardianShield runs on your own machine — the client starts it, so there is no endpoint to ping. 84 installs a week from pypi. Last commit 3 Mar 2026.
AI security layer: code scanning, PII detection, prompt injection, secrets, CVEs
Today is the operative word: we check GuardianShield every 15 minutes and re-read its code on every release. Watch it and you find out the day that stops being true.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add GuardianShield -- uvx guardianshield
{
"mcpServers": {
"GuardianShield": {
"command": "uvx",
"args": [
"guardianshield"
]
}
}
}
[mcp_servers.GuardianShield]
command = "uvx"
args = ["guardianshield"]
{
"mcpServers": {
"GuardianShield": {
"command": "uvx",
"args": [
"guardianshield"
]
}
}
}
{
"servers": {
"GuardianShield": {
"type": "stdio",
"command": "uvx",
"args": [
"guardianshield"
]
}
}
}
Optional settings from the registry entry, left out of the block above:
GUARDIANSHIELD_PROFILE (Default safety profile (general, education, healthcare, finance, children)), GUARDIANSHIELD_AUDIT_PATH (Path to the SQLite audit database), GUARDIANSHIELD_DEBUG (Set to 1 for debug logging).
Code security scanner for founders and small teams: leaked secrets, missing RLS, injection, CVEs
Security scan for AI-generated code: injection, SSRF, secrets, weak crypto, unsafe deserialization.
Dead code, security, secrets detection and code quality for Python, TypeScript, Go.
Scans code for insecure LLM/AI integration: exposed keys, unsafe output, prompt injection
Scans code for insecure LLM/AI integration: exposed keys, unsafe output, prompt injection
GitHub Actions workflow security audit - 21 checks: pinning, permissions, secrets, injection.
Security scanner for AI-written code: SAST, secrets, dependencies, IaC, containers and DAST.
Security audit for AI agents — scan code/diffs for leaked secrets, check deps via OSV.
Answers built from our own checks of this server.