|
Latch — Mission Control for AI Coding Agents
Terminal-first policy enforcement, runtime authorization, and complete audit trails for Claude Code, Codex, and OpenClaw. Run any agent. Govern everything.
|
87 % |
|
ToolHost — Governed MCP gateway for agent tool access
Connect MCP-speaking agents to a focused, governed tool surface. ToolHost helps teams curate tool access, apply policy, and retain operational evidence.
|
86 % |
|
CallLint — before your agent acts, check the blast radius
CallLint statically scans MCP and agent-tool configs and returns SAFE / REVIEW / BLOCK / UNKNOWN with evidence. Offline by default, deterministic, never exec...
|
86 % |
|
Attendant — AI Agent Governance for Claude, Cursor & Codex
Govern every desktop AI agent — Claude Code, Cowork, ChatGPT Codex, Cursor — with inline enforcement, prompt-injection defense, and SIEM-ready audit.
|
85 % |
|
Navil — Production governance for AI agents
Open-source runtime governance for AI agents. A 2.7 µs Rust proxy in front of every MCP server your agents reach. Policy at install, enforcement at every call.
|
85 % |
|
AgentZero — Secure AI agent runtime
Policy-controlled, audited, encrypted tool execution for local AI workflows.
|
85 % |
|
Scopebound — Zero-Trust Enforcement for AI Agents
|
85 % |
|
Enforra | Action governance for AI agent tool calls
A 5-minute SDK to define, test, trace, and enforce what AI agents are allowed to do before tool callbacks create real side effects.
|
85 % |
|
Fendray | The agent security platform
Fendray is the agent security platform. Inspect, govern, and audit every AI agent tool call across LangChain, MCP, and raw provider APIs.
|
85 % |
|
MCP Gateway & AI Agent Authorization | Endram
Control AI agent and MCP tool calls with identity-aware policy, human approvals, one-use grants, and signed evidence.
|
84 % |
|
Tenuo | Task-scoped authorization for AI agents
Give AI agents task-scoped authority, check every action before it runs, and narrow permissions across delegation. Open source, Apache 2.0.
|
84 % |
|
Sekuire - Guardrails and Permissions for AI Agents
Add guardrails, permissions, and runtime policies to AI agents. Open SDK for local enforcement. Managed platform for fleet visibility, registry, and compliance.
|
84 % |
|
AgentWard — Secure every agent action, from install to runtime
Secure every agent action — from install to runtime. Source-available security platform for AI agents. Scan, enforce, detect, comply.
|
84 % |
|
Tandem | Governance runtime for AI agents
AI agent authority runtime for risk-tiered actions, human-in-the-loop approvals, scoped MCP tools, durable pause and resume, and protected audit evidence.
|
84 % |
|
ActaMeshAI | Accountability for Enterprise Agents
Put policy, approvals, scoped credentials, and evidence between enterprise AI agents and the tools they use.
|
84 % |
|
Aegis — Runtime capability enforcement for AI agents
Aegis is a deterministic capability-enforcement layer for AI agents on MCP. No LLM in the enforcement path. Every tool call evaluated against a spec, with a...
|
84 % |
|
Axyon — The control plane for production AI
Axyon gives platform teams one control plane to deploy models, run agents and decide what they are allowed to do, across the Kubernetes clusters you already...
|
84 % |
|
Backplanes: intent-based visibility and control for AI agents
Backplanes turns your business rules into policy your agents have to follow, enforces it while they work, and records the reason behind every decision.
|
83 % |
|
Kelhe — The runtime control point for AI agent actions
Most AI agents never leave the pilot because no one can trust what they'll do in production. Kelhe scores every action your agents take — before it runs — an...
|
83 % |
|
Tria - See and Control AI Agent Actions
Tria shows what AI agents actually do in production, finds risky behavior patterns, and helps teams block unsafe actions before they run.
|
83 % |
Detected social media networks