|
authoxi — identity, permissions and spend control for AI agents
The control plane for AI agents: identity, permissions, usage and money, enforced at every action, with a signed record of who approved what.
|
86 % |
|
ifivo: Agent Control Plane for AI (Policy, Approvals, Kill Switch)
ifivo is the runtime control plane for AI agents. Enforce spend limits, vendor allowlists, and risk policies. Approve high-risk actions from Slack or in-app....
|
85 % |
|
ProveLLM — Permission before execution. Proof after every decision.
Every AI call runs a multi-gate pipeline against its agent's signed passport before it reaches the model — disallowed model, over budget, outside its allowed...
|
85 % |
|
Perisec | AI Gateway and Policy Enforcement for AI Agents
Perisec sits between every agent and every model or tool call it makes, enforcing policy in real time, before an unauthorized action happens instead of after.
|
84 % |
|
Intercis | AI Agent Governance for Security Teams
Intercis sits between your agent's code and the Anthropic and OpenAI APIs and deletes a denied tool call from the response before your runtime can run it.
|
84 % |
|
corri, the open-source auth layer for AI agents
Open-source middleware that verifies a signed AI agent's identity and applies your access policy, allow, deny, or charge, before your protected code runs. Ad...
|
84 % |
|
Superslate | TypeScript SaaS Boilerplate for Coding Agents
SaaS boilerplate and starter kit for the agent era. Production TypeScript with rules your agent follows: agent context, agent tooling, and golden paths.
|
84 % |
|
AI agent permissions, approvals and model limits | ToolBrake
Set boundaries for AI agents. Private beta with scoped keys, model limits, tool approvals and audit history.
|
84 % |
|
DigitalPipe: Let Your Agents Touch Real Systems
DigitalPipe is the control layer for MCP traffic. One endpoint for every tool your agents reach — route it, inspect every call and result crossing it, and ki...
|
83 % |
|
Statio · Grant the tool, not the whole server
An MCP gateway that controls access one tool at a time and gives every AI agent its own identity. Keys stay encrypted in Statio, and every call is recorded.
|
83 % |
|
nonsudo — AI agents that can do anything shouldn't.
nonsudo intercepts every AI agent tool call, enforces your policy before the call leaves, and produces a tamper-evident receipt. The enforcement layer betwee...
|
83 % |
|
Autonomous Agent — Software delivery that proves itself
Independent, evidence-backed verification for autonomous software delivery — not the agent's word for it.
|
83 % |
|
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...
|
83 % |
|
AgentBouncer - Policy gateway, human approvals, and audit log for AI agents
The gateway between AI agents and production. Policy allows, denies, or holds each action for human approval in Slack. Scoped keys, full audit log. Works wit...
|
83 % |
|
Gavrun - A verdict in front of every agent action
Gavrun is the control plane between your AI agents and the systems they touch. Every tool call gets a verdict: allow, hold, or block, with a record by decisi...
|
83 % |
|
Inferova — Identity, Onboarding & Payments for AI Agents
Inferova is the access layer for autonomous agents — let your agents discover, sign up for, authenticate into, and pay for software with no human in the loop.
|
83 % |
|
Obsign: audit proof of AI agent actions | DORA & NIS2
Enforceable audit log of AI agent actions: who delegated what, to which agent, with which policy verdict. Sealed, verifiable offline. For DORA and NIS2.
|
83 % |
|
Talaiz — Secure Access for AI Agents
Secure Access for AI Agents: an audited, outbound-only gateway between MCP clients and your internal systems — deny by default, every call on the record.
|
83 % |
|
Squidder | The control plane for enterprise AI egress
Squidder is one governed doorway for everything your services send out: web, LLM, and agent (MCP) protocols. Deny by default, inspect, tokenize, put the real...
|
82 % |
|
Oberik — add AI to your product without building the hard half
Oberik is the agent layer you drop into your SaaS. Every customer gets answers grounded in their own documents — with page and quote — plus tools, a Linux sa...
|
82 % |
Detected social media networks