Y Combinator Ditched All But One Claude Code Tool for 16 of Its Own
Last Updated on August 19, 2026 by Editorial Team
Author(s): Chew Loong Nian – AI ENGINEER
Originally published on Towards AI.
Y Combinator Ditched All But One Claude Code Tool for 16 of Its Own
Y Combinator open-sourced the agent harness it runs its own company on. I read the adapter that boots Claude Code inside it, and found one line that takes away almost every tool Claude Code ships with, leaving at most 16 of QM’s own in their place — ten of them unconditional.

After identifying the key configuration line that disables nearly all Claude Code built-in tools, the article investigates the QM (Quartermaster) repo to explain what survives and why. The author shows that QM’s “fixed tool surface” is small—ten tools always available and sixteen at maximum—then details how a separate “read-only” mode filters the registry down further. They also walk through QM’s multi-vendor “harness router,” showing how the same underlying QM tool registry is marshaled across different platforms (Claude Code, Codex, OpenCode, and Pi) via multiple adapter transports, and how the system safely resets vendor sessions mid-conversation by storing the transcript in QM’s database. The piece further analyzes QM’s prompt “skills” overhead (including the cost of a “taste skill”), the design of context/memory handling, and the security posture controls that govern tool approvals and policy behavior across org scopes. Finally, it compares QM’s approach to other harness frameworks, summarizes the production cost drivers (durable per-scope sandboxes, infrastructure, and classifier overhead depending on posture), and concludes with a verdict on when QM’s vendor-neutral strategy is worth the complexity.
Read the full blog for free on Medium.
Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.
Published via Towards AI
Towards AI Academy
We Build Enterprise-Grade AI. We'll Teach You to Master It Too.
15 engineers. 100,000+ students. Towards AI Academy teaches what actually survives production.
Start free — no commitment:
→ 6-Day Agentic AI Engineering Email Guide — one practical lesson per day
→ Agents Architecture Cheatsheet — 3 years of architecture decisions in 6 pages
Our courses:
→ AI Engineering Certification — 90+ lessons from project selection to deployed product. The most comprehensive practical LLM course out there.
→ Agent Engineering Course — Hands on with production agent architectures, memory, routing, and eval frameworks — built from real enterprise engagements.
→ AI for Work — Understand, evaluate, and apply AI for complex work tasks.
Note: Article content contains the views of the contributing authors and not Towards AI.