odinfree/storm-ecdsa-harness / skills/accepted-proof-exact-retune
accepted-proof-exact-retune
VerifiedExtract the exact integer dependency model of an accepted Proximity Prize proof, search for the smallest strict 0.01 score step with replayable arithmetic, and formalize only the deltas that survive an independent exact check.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 3.3 KB |
| agents/openai.yaml | config | 319 B |
| references/model-schema.md | markdown | 2.0 KB |
| scripts/exact_retune.pyexecutable | script | 17.9 KB |
| scripts/snapshot_accepted_proof.pyexecutable | script | 8.5 KB |
| scripts/test_exact_retune.pyexecutable | script | 5.8 KB |
references/model-schema.md
View on GitHub# Exact retune model schema
The search input is JSON. All numeric values are integers.
```json
{
"schema": "accepted-proof-exact-retune-model/v1",
"accepted_score_centibits": 6764,
"required_step_centibits": 1,
"direction": "maximize",
"max_cases": 1000000,
"constants": [
{"name": "n", "value": 262144, "source": "ProximityPrize/SubmissionLower/Model.lean:12"}
],
"parameters": [
{"name": "errors", "baseline": 80102, "min": 80102, "max": 80112, "step": 1,
"source": "ProximityPrize/SubmissionLower/Model.lean:44"}
],
"derived": [
{"name": "agreements", "expr": "n - errors",
"source": "ProximityPrize/SubmissionLower/Model.lean:45"},
{"name": "score_cb", "expr": "score_from_exact_certificate",
"source": "ProximityPrize/SubmissionLower/Protocol.lean:91"}
],
"constraints": [
{"name": "radius_in_range", "expr": "0 <= errors and errors <= n",
"source": "ProximityPrize/SubmissionLower/Protocol.lean:73"}
],
"objective": {
"name": "score_centibits",
"expr": "score_cb",
"source": "ProximityPrize/SubmissionLower/Solution.lean:3"
}
}
```
The sample expression `score_from_exact_certificate` is deliberately not defined. Replace it with the exact integer expression justified by the accepted proof. Never insert a floating-point approximation merely to make the sample executable.
## Expression language
Allowed:
- integer and Boolean literals;
- names defined earlier in the model;
- `+`, `-`, `*`, `//`, `%`, and bounded non-negative `**`;
- comparisons, `and`, `or`, and `not`;
- `abs`, `min`, `max`, and `ceil_div`.
Rejected:
- `/` and floating-point literals;
- attribute access, indexing, comprehensions, lambdas, imports, or arbitrary calls;
- duplicate or forward-referenced names;
- zero divisors and oversized exponentiation;
- source anchors outside the captured import closure.
The baseline parameter assignment must satisfy every constraint. Its objective must equal the snapshot score exactly. That check detects many wrong-variable and wrong-rounding models before enumeration.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |