Intel / article

Use ACP-compatible harnesses with the AI SDK harness layer

Source: Felix Arntz
Author: Felix Arntz
Date: Published Aug 13, 2026

Terms in this piece · Glossary

agent harness — The scaffolding around a model that turns it into a working agent — the loop, the tools it can call, and the rules for when to stop.
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.

Why it matters

Anyone building on top of coding agents can now target a protocol instead of maintaining one adapter per runtime. That collapses the integration cost of supporting a new to a small config block.

Key quotes

“The AI SDK harness layer now supports any Agent Client Protocol (ACP)-compatible harness with HarnessAgent through the new @ai-sdk/harness-acp package.”
Felix Arntz

“Previously, every harness adapter wrapped one specific runtime (Claude Code, Codex, Pi, Deep Agents, OpenCode). @ai-sdk/harness-acp wraps the protocol instead.”
Felix Arntz

“ACP is an abstraction over coding harnesses, but the AI SDK harness layer stays deliberately decoupled from it. Not every harness supports ACP, and for some, ACP limits or changes how well they expose their internal behaviors, so a direct adapter can produce a tighter integration.”
Felix Arntz

“For Claude Code and Codex, prefer the direct @ai-sdk/harness-claude-code and @ai-sdk/harness-codex adapters over an ACP-based implementation; reach for @ai-sdk/harness-acp when a harness has no direct adapter but ships an ACP-compatible package.”
Felix Arntz

Read the source vercel.com