Research companies and people hypothesis-first

Test hypotheses and produce decision-grade entity dossiers

Investors and operators who need decision-grade entity research with hypothesis testing and red-flag scoring before meetings or deals.

• · Entity (company, person, nonprofit, government org)
• · User's hypothesis (what they believe upfront and want to verify/disprove)
• · Optional BYOK MCPs (LinkedIn, Crunchbase, Apollo, Pitchbook, SimilarWeb)

• · Editable Word document (.docx) with verified hypothesis verdict
• · Identity facts (legal name, leadership, location, funding)
• · 12-month activity timeline (key events, milestones)
• · Network signals (connections, partnerships, co-mentions)
• · Reputation signals (media mentions, awards, controversies)
• · Red flags (structured risk factors)
• · 3-5 conversation hooks tied to specific findings
• · Source-provenance audit log

• · WebSearch + WebFetch (free)
• · Free APIs: SEC EDGAR, GitHub, ProPublica Nonprofit Explorer
• · Optional BYOK MCPs: LinkedIn, Crunchbase, Apollo, Pitchbook, SimilarWeb

• · User must state hypothesis upfront (verified vs confirmed bias resistance)
• · Entity must be identifiable by name or URL
• · Public information must exist (dossier is for public entities, not private persons)

• · Hypothesis confirmation bias can bias search (mitigated by forcing upfront statement)
• · Red flags may be over-weighted if reputation sources are limited
• · Conversation hooks may miss personal relationship history (public signals only)

• · Source spec in PR #657 (megaprompts/12-dossier-megaprompt.md)
• · Hypothesis-testing variant forces upfront statement (bias resistance)
• · Free APIs (SEC EDGAR, GitHub, ProPublica) are production integrations
• · DOCX output is editable and source-tracked
• · Conversation hooks are tied to specific findings, not generic talking points