AgentPort
Open-source security gateway for AI agents — granular tool policies, approval chains and full audit logs
0Visit AgentPort
agentport.shAbout AgentPort
AgentPort is an open-source security gateway that sits between AI agents and the tools they call, so an agent's access to production systems is governed by policy rather than by whichever API keys happened to end up in its environment. Agents connect through either an MCP server or a CLI, and from there reach Linear, Slack, GitHub, Notion, Stripe and dozens of other connectors that ship out of the box, plus any additional MCP server or REST API you point it at. The control layer is where the value sits: you define which tools a given agent may call outright, which require a human approval before execution, and — for the more cautious — advanced rule-based filtering that can gate on the shape of the call rather than just its name. Every tool call and every approval chain is written to an audit log, which turns an agent's behaviour from something you infer from outcomes into something you can read after the fact. This is the missing piece in most agent deployments: the agent framework decides what to attempt and the credential decides what is possible, with nothing in between. Deployment is either self-hosted from the open-source repository on your own infrastructure, or a managed cloud service currently in beta. AgentPort is built by Skald Labs, based in San Francisco.
Does ChatGPT recommend your AI tool?
If you're building in AI Agent Infrastructure, run a free AI-visibility scan on your own product — we ask ChatGPT across 5 prompt angles and score how often you get named. ~30 seconds, no signup, no card.
Key Features
AgentPort Pros & Cons
✅ Pros
- +Policy-plus-approval between agent and credential is the missing layer in most deployments
- +$12/mo for unlimited integrations is unusually cheap for a security control plane
- +Open source means the gateway itself can be audited
⚠️ Cons
- −Cloud service is still in beta
- −Human approval in the loop defeats the point for fully autonomous workflows
- −Team-level policy management is gated behind unpriced Enterprise
Tags
Is AgentPort your tool?
This is the page buyers and AI assistants read when they look up AgentPort. Claim your listing for $19 one-time — no subscription, nothing to cancel — and get a Featured badge, top placement in your category, and a permanent dofollow backlink. Prefer it ongoing? Monthly is one click away on the next page.
Complete Your AI Agent Stack
Other ai agent tools in our catalog:
1Password
Try FreeSecrets and credential manager
Keep API keys and .env secrets out of your repo
Consensus
Try FreeAI search across 200M research papers
Source real evidence behind your analysis
Gamma
Try FreeAI presentation builder
Turn ideas into polished decks instantly
💰 Affiliate disclosure: We may earn a commission if you sign up through these links at no extra cost to you.
Stay updated on AI Agent Infrastructure tools — join our weekly newsletter
One concise email with fresh launches, trending picks, and featured standouts.
Alternatives to AgentPort
View all AgentPort alternatives →More AI Agent Infrastructure tools
l6e
An open-source MCP budget gate that gives a coding agent a dollar limit per session — cheaper runs and tighter scoping, free via pip
machine0
Persistent NixOS and Ubuntu VMs built for long-running agents, driven entirely by CLI and MCP, billed per minute from $0.013/hr
Maritime
One micro-VM per AI agent with per-customer isolation, lifecycle webhooks and a non-interactive CLI — 3 agents free, then $20/mo for 20
Agentset
Open-source RAG platform for production AI chat and search — multimodal ingestion, automatic citations and metadata filtering behind JS and Python SDKs.
agentskills.codes
Discover and install trusted Agent Skills for Claude, Codex, Copilot, and other AI agents.
AgentsProof
Trace, grade and share AI agent eval reports from a few SDK calls, with custom graders and golden cases
Agent connectivity: not yet verified