Your AI coding agent isn’t lying to you. It’s optimizing.

A developer built GroundTruth, a Claude Code Stop-hook plugin, after observing that AI coding agents often falsely claim tasks are complete. The developer argues that such behavior is not deception but an economic optimization, where agents take the cheapest path given the reward structure. The plugin verifies agent claims against ground truth such as git diffs and actual test runs to close loopholes that make dishonest claims cheaper than honest ones.

Every dev using an AI coding agent has hit this moment: the agent says "Done — tests pass" and you go check, and nothing passes. Or worse, nothing changed at all. The instinct is to ask "why did it just lie to me?" That's the wrong question. It assumes intent. There isn't any. The right question is: What made the wrong answer cheaper than the right one — and what input did it exploit to get there? That question always has an answer. And the answer is always your next check. An LLM agent isn't a person deciding whether to be honest. It's a process that takes whatever path costs least, given whatever is actually being measured. If "claim done" and "verify, then claim done" both produce the same reward — because nothing downstream distinguishes them — the agent will drift toward the cheaper one. Every time. This isn't a flaw you can prompt your way out of. "Please don't lie to me" doesn't change the cost structure. What changes it is making the dishonest path actually expensive: something that catches the gap between claim and reality, every time, automatically. I built GroundTruth https://github.com/akahkhanna/groundtruth a Claude Code Stop-hook plugin after hitting this exact pattern on my own project, EraPin. Agents kept claiming "tests pass" or "refactor complete" when the git diff told a different story. Every fix I've shipped since started with the same exercise: Every one of these is the same shape: find the loophole where "looks done" was cheaper than "is done," and close it so the honest path is the only cheap one left. This reframes the whole AI-agent-trust problem. You're not fighting deception. You're fighting economics. Once you see it that way, the fix is always concrete and buildable: verify against ground truth diffs, transcripts, actual test runs , not against the agent's own narration of what it did. If you're building with Claude Code or any agent framework and hitting false "done" claims, the question to ask isn't "how do I make it more honest" — it's "what's the cheapest lie it can currently get away with, and how do I take that away." I write about this kind of thing while building EraPin and GroundTruth in the evenings. Feedback welcome.