Map / AI Endpoint Security / Full Court Defense

Full Court Defense

Israel

Quick answer: Full Court Defense is an AI security company based in Israel, working in AI Endpoint Security, MCP & LLM Gateways, Observability & Governance and 1 more category. It was previously known as BotGuard.

AI Endpoint SecurityMCP & LLM GatewaysObservability & GovernanceRuntime & Guardrails

Capabilities

Per-machine agent and CLI that checks every action an AI coding tool or agent takes before it runs: shell commands, file edits, MCP tool calls and API calls. Each rule can allow, ask the developer, require manager approval, block, or mark an action never allowed, with a Monitor-only mode to start. Inventories AI tools, agents, MCP servers, agent config files and exposed secrets on every enrolled machine, adds behavior detection on unusual action sequences and a posture scan of AI tool settings. Python and Node.js SDKs apply the same action policies to tool calls in agents built with LangChain/LangGraph, OpenAI Agents SDK, Google ADK, CrewAI, AutoGen, LlamaIndex and Semantic Kernel; cloud connectors build an AI-BOM.

Sensors / integration

One install per machine (Windows, macOS, Linux) on laptops, VMs, servers and CI. Hooks inside Cursor and Claude Code intercept commands, file writes and prompts; a local MCP proxy rewrites MCP client configs for Cursor, Claude Code, Claude Desktop, Codex, Gemini, Windsurf and VS Code. Python and Node.js SDKs wrap tool calls in custom agents. Optional connectors pull cloud AI inventory from AWS Bedrock, Azure AI, Vertex AI, OpenAI, Anthropic, Slack, Agentforce and Copilot Studio.

Protections

Blocks, or holds for developer or manager approval, risky agent actions such as destructive commands, database writes, outbound email or payment API calls, and disallowed MCP tool calls. Decisions are made locally and keep working offline (fail-open except never-allowed actions). Flags anomalous agent action sequences, finds secrets AI tools can read, and masks secrets and medical data on the machine before data leaves. Every decision is logged with the matching rule and exported to SIEM (Splunk HEC, Elastic ECS, OCSF).

Differentiator

On-endpoint, pre-execution policy enforcement across both commercial coding agents and in-house SDK agents, with human-approval workflows. Rebranded in 2026 from BotGuard, a chatbot red-teaming and LLM-firewall tool. Small team led by CEO Nahum Segal, CTO Boaz Lautman and COO Lior Levy (ex-IDF cyber defense operations, former CISO); no outside funding disclosed; pricing not public.

Team & funding

Founders
Nahum Segal, Boaz Lautman, Lior Levy
Previously
BotGuard

Other AI Endpoint Security companies

Anomity · AstraSync AI · Asymptote Labs · Autonomous · Backslash · Bay Security · Bloom · BlueRock · Bold · Canyon Road · Capsule Security · Certiv · See all →

Last updated · Get the “Listed on” badge