Research Proof

Forces research and benchmark claims through a frozen-verifier proof ladder instead of confident prose.

par tonyblu331 · tonyblu331/research-proof

Testé · Fonctionne ★ 9.2/10

Research Proof — Forces research and benchmark claims through a frozen-verifier proof ladder instead of confident prose.

Ce que fait

Turns a vague research, benchmark, or causal claim into a structured proof program with fixed Claim / Verifier Boundary / Rejection Gates / Verdict / Proof Ledger headings. Triggers on research reviews, eval design, benchmark or causal claims, and evidence-certainty questions, and forces PROVEN/SUPPORTED/REJECTED/OPEN labeling instead of unqualified confidence.

Rapport de test

Ran a real production-latency claim through both paths: the skill version forced an independent-evaluator check and a randomization rejection-gate the freeform baseline never raised, then correctly downgraded the verdict from an informal 'strongly supported' to the disciplined SUPPORTED label.

Testé le: 2026-07-14 · Claude Code 2.x (agent harness)

Installation

git clone https://github.com/tonyblu331/research-proof.git
cd research-proof
mkdir -p ~/.claude/skills
cp -r skills/research-proof ~/.claude/skills/research-proof

Commandes et exemples de prompts

/research-proofForces research and benchmark claims through a frozen-verifier proof ladder instead of confident prose.

Les skills se déclenchent sur des demandes en langage courant — aucune commande à retenir. Après installation, des prompts comme ceux-ci l'activent (en anglais) :

Prove this latency benchmark claim before I publish it
My research says X is faster — verify that properly
Check whether this experiment actually supports its claim

Rapport complet →