What humans do is keep moving from imperative to declarative in programming languages, and in how agents are prompted via goals and loop definitions.
- The essence of an LLM loop isn’t “autonomy”; it’s “verifiable iteration.”
- It’s not that the agent is smart and will eventually solve it if you let it run.
- You need a mechanical way to judge success/failure at every iteration.
- In a good loop, the exit condition matters more than the prompt.
- You need a clear stop signal like “tests pass,” “compile succeeds,” “metric improves,” or “zero exit code.”
- Without an exit condition, a loop is closer to gambling than engineering.
- Separate what the agent decides from what the code must enforce.
- LLM: ambiguous judgment, code generation, diagnosis.
- Deterministic code: execution, verification, logging, retry caps, rollback, diff checks.
- An agent loop is ultimately a software architecture problem.
- Loops where an agent evaluates itself are risky.
- It can keep rationalizing within the same flawed logic.
- It’s safer to separate the task agent and the evaluator agent.
- The biggest risk isn’t cost; it’s humans losing understanding.
- The faster an agent produces code, the less people know “why it ended up this way.”
- When a production failure happens later, debugging becomes very difficult.
Matt Van Horn on Twitter / X
https://t.co/DM0CAuyprS— Matt Van Horn (@mvanhorn) June 8, 2026
https://x.com/mvanhorn/status/2063865685558903149
The loop is plumbing. The asset is the skill it calls.
Principles beat rules
If someone mentions a bug, say this. If someone compares us to another tool, say that. If someone asks about pricing, mention this plan. This was very brittle.
- Be helpful, not defensive.
- Do not talk down to the user.
- Check factual claims against the docs.
- Sound like someone who builds the product, not someone who processes feedback.
This made the skill file smaller and the agent a lot better.
Petra Donka on Twitter / X
https://t.co/kI2B1isN4N— Petra Donka (@petradonka) May 14, 2026
https://x.com/petradonka/status/2054897826149101588

Seonglae Cho