PromptPotter is LLM-driven program evolution for prompts and pipeline parameters. Point it at your backend, run one command, walk away — the Potter critiques, regenerates, and converges on a prompt that holds up under measurement.
Every round, L1 reads the prior winners and the critique log, then proposes a fresh population of candidate prompts. PoBB elimination (ε=0.05, n_min=4) culls the weak so spend tracks signal — no flat sweeps, no wasted queries.
round_0004/candidates.json winner_id "6a674185" composite 0.94 // +0.31 vs origin parent "3c1d9020" (round 3 winner) eliminated 11 of 16 // PoBB ε=0.05, n_min=4 critique "add explicit combinatorial count cue"
When L1 plateaus across rounds, L2 fires once. It reads the stalled critique stream and rewrites task_context — never pipeline_params. The next L1 round inherits the new framing and goes again, often with a step-change in fitness.
// L2 fired after 3 stalled rounds on hard samples task_context "This dataset rewards step-by-step combinatorial enumeration. Candidates that skip the count step consistently miss multi-hop entries — make the counting explicit before the answer." pipeline_params (unchanged — L2 never edits these)
L3 fires only when both L1 and L2 have stalled. It scraps the plan and writes a new one — different sample budget, different scoring composite, different exploration policy. The new plan becomes the contract for round five and on.
round_0007/l3_plan.json verdict "replan" budget samples_per_round: 64 → 128 composite accuracy + latency_penalty * 0.15 policy "prefer wide exploration over fine sweep" rationale "L2's task_context edit didn't move composite past 0.71 — the scoring shape is wrong, not the prompt"
promptpotter connector/ui polls every two secondsSix datasets, each with its own failure mode. Open one to read the campaign log, the critique trail, and the prompt that won.
A thirty-minute path from pip install to your first converged campaign, in plain language.
Why L2 fires only after L1 stalls, why L3 is the last resort, and what the recursive L4 layer is for.
BBEH, AIME, GSM8K, JustLogic, HotpotQA — with protocols, with held-out splits, with the receipts.
“It fixed a six-month-stuck pipeline in an afternoon. Then it just worked.”
— Maintainer dogfooding TermNorm, sprint 14
Free. One command. Read every round in your editor as it lands.