Other · Local MCP server

Agent Guard

Safety checks an agent runs before it acts: package malware, destructive command, secret leak.

What the MCP Registry states

The entry as published to the official MCP Registry (read 2026-10-04), latest version.

Registry name
io.github.ipezygj/agent-guard
Version
0.2.1
Status
Active
Category
other
Transport
stdio (local process)
Package
PyPI
Published
2026-07-23
Updated
2026-07-23
Publisher
ipezygj (GitHub) · 3 servers with pages here
Namespace
Namespace verified by the MCP Registry via GitHub (github.com/ipezygj) · how
Repository
github.com/ipezygj/agent-guard
Source
Registry API entry

Packages

RegistryPackageVersionTransportEnvironment variables
PyPIagent-tripwire0.2.1stdioNone declared

How to connect Agent Guard

Agent Guard runs locally from a Python package published to PyPI: agent-tripwire version 0.2.1. It speaks MCP over stdio, so the client starts it as a program and talks to it through standard input and output. It needs Python; clients usually start it with uvx (from uv) or after pip install - the usual command is uvx agent-tripwire.

In the mcpServers JSON format that many desktop and editor MCP clients read, the entry looks like this (placeholders in angle brackets):

{
  "mcpServers": {
    "agent-guard": {
      "command": "uvx",
      "args": [
        "agent-tripwire"
      ]
    }
  }
}

Derived from the registry entry, not tested here. What the server does, and on what terms, is set by its publisher; check its repository or website before giving it access to your accounts or files. How to add an MCP server to an assistant · Before you connect

More from ipezygj (GitHub)

All 3 →
ServerRunsEndpoint or package
EvalgateStatistical checks an agent runs before trusting an AI eval number (is #1 real, judge bias, more).Local · stdioPyPI: eval-integrity
NumguardVerify a number before an agent asserts it - evals, backtests (Deflated Sharpe), signed receipts.Local · stdioPyPI: numguard

More other servers

All 8,395 →
ServerRunsEndpoint or package
Agent GateLocal mandate gate and proof trail for AI agents that take consequential actions.Local · stdionpm: forge-agent-gate
Agent GateFail-closed quality gate and hash-chained receipt ledger for AI agent workflows.Local · stdioPyPI: mcp-agent-gate
Agent GatewayRelux Works studio gateway: read services/pricing and submit a project inquiry to hire the team.Remote · HTTPLive check: ✓ handshakeapi.relux.works
Agent Graph AuditorFinds transitive excessive agency across LangGraph/CrewAI/AutoGen orchestration graphs.Local · stdionpm: @guardbee/mcp-agent-graph-auditor
Agent Guard MCPVerify-before-act safety checks for AI agents: packages, lockfiles, manifests, CI workflows.Local · stdionpm: @liminallablibs/agent-guard-mcp
Agent GuardrailDeterministic policy firewall for AI agent tool calls - YAML rules, not a fuzzy risk score.Local · stdioPyPI: guardrail-mcp
Agent Guardrail MCP ServerAction-level governance for AI agents -- control what they DO, not what they SAY.Local · stdioPyPI: agent-guardrail
Agent Hive OrchestratorDeterministic nested hive-mind orchestration over the A2A rails: hire N solvers, adversarial cross-r.Remote · HTTPLive check: ✓ handshakemcp.viridisconservation.com