compare-creators

name	compare-creators
description	Compare two or more accepted Creator Dossiers across positioning, audiences, values, content portfolios, performance tiers, proof systems, directing and visual patterns, evolution, demand, and commercial evidence. Use for multi-creator research; do not turn comparison into copying advice.

Compare Creators

Compare accepted, version-pinned Creator Dossiers. Preserve non-comparability instead of normalizing unlike evidence into a false ranking.

Inputs

• two or more creator IDs;
• fixed dossier run IDs and revisions;
• their readiness/gate reports and creator evaluation reports when available;
• requested comparison questions.

Method

1. Verify each dossier revision and evidence health. Quarantine failed sections.
2. Build a comparability matrix for metric coverage, snapshot era, corpus size, tiers, deep-video coverage, comment scope and platform context.
3. Compare positioning, audiences, values, trust, lifecycle, topics, formats, proof, directing, visual/editing, publishing evolution, public demand and commercial evidence.
4. For every shared pattern record supporting creators/videos, qualifying conditions, contradictions, confidence and unknowns.
5. Separate shared observation, conditional mechanism, and creator-specific exception.

Agent-loop boundary

Run every creator through the canonical single-creator loop independently; one blocked member must not stop other members. The batch owns ordering, aggregate progress, failure isolation, and fixed revision references, but never a second mutable copy of creator evidence.

Only start comparison after the requested creator revisions are pinned and the comparability matrix is available. If a comparison fails, classify whether the cause is a missing creator artifact, incompatible denominator/snapshot/platform, unsupported cross-creator claim, or evaluator finding. Rerun only the affected member or comparison node after its input, evidence, contract, or code fingerprint changes. Never restart ready creator runs merely to make the batch look complete.

Independent comparison evaluation is optional for a working comparison and required before formal Wiki promotion. Findings remain visible and do not trigger automatic repair.

Produce creator-comparison.json with pinned sources and stable evidence links. Do not recommend what the user should copy, produce, title, or publish. Pass accepted comparison evidence to $content-strategy-workbench only when the user starts a separate creation task.

Return COMPARISON_READY only when every displayed comparison is supported and source revisions remain current. Return COMPARISON_WIKI_READY only when the pinned comparison revision also passes independent promotion evaluation.