OpenClaw: Securing Agent Swarms

Learn to secure agent swarms by enforcing policies on tool use, including filesystem access, command execution, and network egress, with practical detection and audit trails.

Overview

Agentic coding isn’t “one process on one host” anymore. It’s a swarm of local and hosted planners/coders/reviewers using tools that touch files, run commands, and egress to the internet. The security boundary isn’t the prompt; it’s the tool boundary where intent becomes action.

In this session I’ll do a walkthrough of a practical “SDR” layer for OpenClaw-style workflows:

Policy enforcement on filesystem + command execution (path allow/deny, traversal hardening, safe defaults)
Network egress control (where requests can go, CONNECT/TLS sanity checks, blocking risky sequences)
Signed receipts / audit trails so you can replay, diff, and prove what actually happened during a run
A small demo of writing a policy and watching it block + record tool calls in real time

Links

Tech stack