Build your own AI agent harness: memory, permission gates, and subagents
PawelHuryn · x · 2026-11-05
Developer Pawel Huryn shares how to build a full agent harness from scratch: starting from a working agent, readers can reverse-engineer his GitHub repo to fill in the rest.
His newsletter walks through the build step by step with copy-paste prompts, covering:
- Sessions, compaction and memory: an agent that survives crashes and long tasks
- A permission gate plus an AI judge: nothing risky runs without a yes
- Subagents and goal mode: work checked by something other than the worker
- Skills, MCP servers and cost: what plugs in and what it costs
He previously audited competing small harnesses against their own source code, claiming only his covers all common features.
Related event: Developer Open-Sources 1,200-Line Teaching Agent Harness in Python(4 posts)→
More from coding & agent
- tw93 open-sources Kami: a design system that makes AI-agent documents actually readable — sujingshen · 2026-11-05
- Power user shares the one thing he won't let Claude do: write his tweets — PrajwalTomar_ · 2026-11-05
- Old iPad reborn as a smart home wall panel, thanks to Claude — PrajwalTomar_ · 2026-11-05
- He replaced paid app subscriptions with Claude-built tools, saving $200/month — PrajwalTomar_ · 2026-11-05
- Running Claude Code on DeepSeek for chores: one job costs $0.0043 — PrajwalTomar_ · 2026-11-05
- Agent audited a 300K-star skill pack in 70 seconds, found startup hook and usage counter — PrajwalTomar_ · 2026-11-05