v0.2.0 - Rigor pipeline

Ignat v0.2.0 — rigor pipeline: anti-slop gate, formal proofs, manuscript architect.

This release extends Ignat from review toward a full rigor pipeline — is it well-argued? → is every fact real? → which claims can we prove outright?

New commands

• /groundcheck — the AI-slop gate. Where /review judges the whole paper, /groundcheck walks every load-bearing claim and asks "is this grounded in a verifiable source?" Catches the five things models hallucinate most: vibe citations, ungrounded numbers, entity/property confusion, dangling cross-references, and unverified "first/novel" claims. Agent-driven — no extra tooling.
• /formalize — turn the core claims of a paper into machine-checked theorems (Lean 4 + mathlib / KeYmaera X), with honest scoping (machine-checked vs pen-and-paper vs unproven) and #print axioms verification (no sorryAx).

New agents

• paper-architect — manuscript strategy director: scope discipline, novelty-honesty, main-vs-SI split, journal fit. Orchestrates the panel; does not write the science.
• verifier — citation & number fact-checker; the dedicated defense against hallucinated references (also resolves a dangling reference in the previous README).

Docs

• docs/SCIENTIST_BEST_PRACTICES.md — the hard-won discipline behind these tools: an agent's "PASS" is a hypothesis, not a verdict; verify the load-bearing number yourself; verify what you computed is what you claim; don't fabricate; honest scope beats an over-reached proof.

Full changelog

v0.1.0...v0.2.0