1. Claim Map
Target material is split into discrete claims, citations, assumptions, and rhetorical maneuvers before a response is drafted.
Prototype site
Rapid claims deserve recorded answers.
This prototype presents an evidence-first response format for antievolution and science-denial claims. The goal is not volume for its own sake. The goal is to make every claim traceable, every citation checkable, and every published response reviewable.
Method
Designed to collapse the Gish Gallop into reviewable units.
2. Corpus Check
Claims are matched against TalkOrigins, the Index to Creationist Claims, Panda's Thumb, TalkDesign, evo-edu, and reviewed source archives.
3. Citation Audit
References are parsed, verified, and checked against source text when available. Misquotation and unsupported deployment are tracked separately.
4. Review Gate
AI drafts are not publication. Reviewers approve source use, factual claims, rhetorical labels, and correction notes before release.
5. Burden Check
Unsupported assertions are marked as unsupported. Hitchens' Razor applies: what is asserted without evidence can be dismissed without evidence.
Demo docket
Candidate response style
"Dances With Popper": An Examination of Dembski's Claims on Testability
This demo uses a legacy Panda's Thumb post as a model response: source quotation, claim isolation, close reading, Index alignment, and citation-aware scrutiny.
Knowledge base
What the workbench will draw from
Foundation corpora
TalkOrigins Archive, Index to Creationist Claims, TalkDesign, Panda's Thumb, Panda's Thumb MT, and evo-edu resources.
Source archives
Books, PDFs, archive.org snapshots, opposition-site captures, author/source histories, and reviewer-supplied materials.
Tooling
GroundRecall for provenance, CiteGeist for citations, doclift for document normalization, Didactopus for evidence paths, Wolfe for dev-host retrieval trials.