In this YouTube commentary, Robert Ta explains Anthropic’s article “A Harness for Every Task: Dynamic Workflows in Claude Code.” He describes how Claude can automatically generate JavaScript harnesses that spawn and coordinate isolated sub‑agents—separate Claude instances that can run with different models (Sonnet, Opus, etc.) and their own work trees. By moving planning and execution into distinct sub‑agents, the system avoids the so‑called “agentic laziness” where Claude declares a task finished despite incomplete work. Ta walks through the mechanisms for triggering workflows (the ultra‑code keyword, loop and goal commands), setting token‑budget limits in plain English, and persisting or sharing workflows via the .claud/workflows directory and Skills. He cites a real‑world migration of Bonn’s codebase to Rust as an example, noting the cost and time involved. Finally, he outlines the six workflow patterns Anthropic recommends: (1) classify‑and‑act, (2) fan‑out‑and‑synthesize, (3) adversarial verification, (4) generate‑and‑filter, (5) tournament‑style judging, and (6) loop‑until‑done. The commentary mixes technical explanation with informal critique, especially of the term “agentic laziness.”
Chapters
Anthropic published this article, A Harness for Every Task: Dynamic Workflows in Claude Code. The thesis verbatim is that Claude writes its harnesses on the fly, custom built for the task at hand. So how this works is Claude writes a JavaScript file with special functions that spawn and coordinate subagents, isolated Claude instances with their own context windows and focus goals. The workflow decides which model each subagent uses, whether it be Sonnet or Opus or whatever, and it could run subagents in isolated work trees. The real problem this solves for AI builders is when you ask the default Claude Code Harness to do a task, it needs to both plan and execute in the same context window. Workflows split those across isolated subagents and combat agentic laziness, is what they say. When Claude stops before finishing a particularly complex multi-part task and declares the job done after partial progress. That's sort of funny that they call it agentic laziness instead of just calling it a lie, which is what it is. It's like Claude, if you're telling me you're done and then I look at your work, it's not fucking done, bro. You, it's not agentic laziness. This is really interesting. How do you actually drive it? Say, use a workflow or the trigger word ultra code to force one. And that combines with the loop command and the goal command for a hard completion target with a keep going sort of command. And you could also set token budgets in plain English, like only use 10,000 tokens, only use 500,000 tokens kind of thing. It saves workflows in the .claud/workflows directory., or you could share them via skills. Some real use cases that have used this, Bonn recently did a rewrite to Rust, and that was a large migration refactor. I think that cost them a few thousand dollars over a couple weeks on a long horizon task. There's 6 new workflow patterns that the Anthropic team recommends with this harness on the fly feature. Number 1, classify and act task goes into a classifier that routes to different agents. Number 2, fan out and synthesize. Task goes out, it fans out, does stuff, come back, and then collates it, synthesizes it. Number 3, adversarial verification. Many verifiers and then running a worker. Generate and filter. Generate stuff, get to ideas, filter out with a rubric, deduplicate ideas, and then do a best of breed amongst them and discard stuff against your rubric and filtering. Tournament style, where you have attempts to do workflows, and then you pair them with judges, which are decision makers. It looks like you can have different tiers of judges here, and then a final judge to determine a winner from those attempts at workflows. Finally, number 6, loop until done, where the agent does stuff and there's a checkpoint, and it'll say, do you have new findings? No. If no new findings, done. If yes, new findings, spawn another agent, keep going until done.
