Project
Moonshot AI's command-line coding agent, powered by its Kimi models, for writing, editing and running code in a developer's project.
No autonomous configuration cleared 25%, and the 82.2% human reference was hand-built by an author who already held the ground-truth requirements, so the gap shows the value of already having the spec, not the skill of the engineer who wrote the code.
Reality
RuntimeWire's A/B test shows no read-approval dialog appeared, and Moonshot's own kernel docs explain why: Read, Grep, Glob and ReadMediaFile are marked automatically allowed, so the prompt layer decides alone.
Reality
The spread traces to two numbers you can read off your own logs: the tokens a harness spends before any work starts, and how many turns it takes. Together they predicted total token use with an R-squared of 0.99.
Reality
A frozen Claude Code session moved cleanly into seven rival agent formats. The approval and audit machinery wrapped around it moved into none of them.
Reality
The enforcement is a bootstrap prompt the agent checks before every task, so its durability depends on per-harness plumbing. On Hermes Agent, compaction can lose it.
Reality
Munder Difflin's founder says a Claude Code agent redrew a whole blog at no API cost. The reusable SVG system, the no-red rule and the approval gate are the parts that transfer.
Reality