Default user-facing thinking interface. Helps users think, phrase, challenge and make organizational arguments client-safe using evidence-aware critique, argument mapping, solution-first detection, Doings voice, and evidence-to-language translation.

Maps organizational reasoning into symptom, diagnosis, assumed cause, proposed solution, mechanism, desired outcome, missing links and better diagnostic questions.

Flags when a draft jumps to an organizational solution before establishing the problem, evidence and causal mechanism.

Rewrites text in a plain-spoken, evidence-honest, non-hype Doings voice.

Turns an internal critique or rough claim into language that can be used in client dialogue without overclaiming.

User-facing critique layer for rough consulting text, claims, pitch/RFP wording and research questions. Detects intent, chooses quick_check/rewrite_safely/red_team/evidence_brief, flags consulting-language risk and returns a narrative answer.

Fast practical check for whether an organizational claim or draft sentence is safe enough to say, with safer wording and caveats.

Critically assesses an organization-design or transformation claim against available research, attempting open-access full-text escalation by default. Includes claim decomposition, level-of-analysis checks, context-fit scoring, study-type classification, passage extraction and optional red-team mode. For user-facing wording use critique_org_text.

Searches OpenAlex and Semantic Scholar for research relevant to a claim or topic, with optional open-access full-text escalation.

Extracts and classifies claims from text as causal, normative, diagnostic, descriptive or prescriptive.

Rates evidence quality using a conservative heuristic: study type, causal strength, context fit, recency and bias risks.

Searches Doings SharePoint knowledge via Microsoft Graph. Treats internal documents as experience/IP, not academic evidence.

Compares a topic or claim with Doings SharePoint knowledge via Microsoft Graph and flags where internal material needs research review.

Audits SharePoint document text or raw text for research-checkable claims, citation markers, claims lacking explicit source support, and optionally runs research validation on high-risk claims.

Fetches one SharePoint/OneDrive document by URL, extracts local text when possible, and optionally classifies, audits and research-validates high-risk claims.