1 min read

Agent-Native Tools Series #9: MCP, Skills, and Tool Contracts

MCP servers, skills, and CLIs matter insofar as they provide dependable tool contracts. Stable schemas, explicit side effects, embedded guardrails, dry runs, and recoverable errors let agents discover, execute, verify, and safely resume real work.
2 min read

Harness Engineering Series #2: Context Engineering Is the Real Interface

A concrete operator guide to context engineering is the real interface: what changes, who owns it, how to verify it, and where it breaks.
2 min read

Harness Engineering Series #1: Why Harnesses Matter

A concrete operator guide to why harnesses matter: what changes, who owns it, how to verify it, and where it breaks.
1 min read

Agent-Native Tools Series #10: The Agent-Native Software Company

An agent-native software company goes beyond adding a chatbot to design its product, pricing, and operating model around delegated work.
2 min read

Harness Engineering Series #4: State, Memory, and Checkpoints

A concrete operator guide to state, memory, and checkpoints: what changes, who owns it, how to verify it, and where it breaks.
2 min read

Harness Engineering Series #3: Tool Design for Agent Work

A concrete operator guide to tool design for agent work: what changes, who owns it, how to verify it, and where it breaks.
2 min read

Harness Engineering Series #6: Human-in-the-Loop as Runtime Design

A concrete operator guide to human-in-the-loop as runtime design: what changes, who owns it, how to verify it, and where it breaks.
2 min read

Harness Engineering Series #5: Orchestration Beyond Chat

A concrete operator guide to orchestration beyond chat: what changes, who owns it, how to verify it, and where it breaks.

Choose your reading rhythm.

Start with a weekly briefing, add daily notes, or hear only when a durable essay or research update is ready.
You've successfully subscribed to Antoine Buteau
You've successfully subscribed to Antoine Buteau
Welcome back! You've successfully signed in.
Press ESC to close.