Invigilate
Your agents, under control
Invigilate is a middleware proxy that sits between an AI model and its external tools, exposing a single URL through which all tool calls are routed. It intercepts each request, allowing users to filter which tools are accessible, enforce custom security policies, and sanitize input and output schemas before they reach the model or the tool. The proxy also records every interaction, generating audit logs that capture the details of each tool invocation.
The system is designed for developers who need to manage and monitor the behavior of AI agents without modifying the agents’ code. By handling security, filtering, and logging at the proxy layer, it simplifies the integration of safety controls into existing workflows. Because it operates as an MCP (Model‑Control‑Proxy) component, it can be inserted into any compatible AI stack with minimal configuration.
Invigilate is positioned as an experimental tool aimed at providing a unified point of control for AI agent tool usage, offering schema sanitization, rule enforcement, and comprehensive auditability in a single, configurable endpoint.
Reviews
Loading reviews…
Similar apps
AI Coding Agents
Korven
AI agents can act. But they have zero security

AI Coding Agents
Tenet AI
Stop tracing AI errors. Start proving your AI decisions.

Password & Security
Guardrly
Monitor AI Agent API calls & prevent account bans.

Security & Identity
QuiGuard
Selfhosted proxy; scrubs secrets from AI Agent tool calls.

AI Coding Agents
Kōan
See your AI agents think. Reasoning, tool calls & decisions

AI Coding Agents
OrchestrAI
Trust every line of AI-generated code. OrchestrAI orchestrates testing, security, quality, compliance, docs and release agents in one…