Axomap
All recipes

Build Landscapes

Create a Landscape from a Thesis

Pro

What this recipe does

Start from a user's market thesis, clarify the scope when needed, research companies that fit the thesis, and create an Axomap landscape after confirmation.

When to use it

Use this when you want to build a new landscape from a thesis instead of starting from a company, deck, or competitor list.

Prompt

Use Axomap's recipe for creating a landscape from a thesis. My thesis is: AI agents that help finance teams automate month-end close.

Expected result

  • A clarified thesis that is specific enough to guide research.
  • A proposed landscape title and concise description.
  • Suggested categories that reflect the thesis boundaries.
  • A selective list of companies that strongly fit the thesis.
  • After confirmation, a new Axomap landscape with categories and companies added.

Requirements

  • The user has a market thesis, investment thesis, category hypothesis, or research idea but no seed company or deck.
  • The user has a ChatGPT Pro or equivalent MCP client account with web research enabled.
  • The user connected Axomap MCP with write access.

Agent workflow

  1. Capture the initial thesis

    Read the user's thesis exactly as provided. It may be a short sentence, a broad hypothesis, or a detailed market description.

  2. Clarify the thesis if needed

    If the thesis is too broad or ambiguous for focused company research, ask a small number of targeted questions about buyer, workflow, geography, company stage, technology boundary, excluded categories, or must-have criteria. Do not over-interview the user if the thesis is already clear enough.

  3. Restate the research basis

    Before researching, concisely restate the clarified thesis and the inclusion/exclusion boundaries the agent will use. This becomes the research basis for company discovery.

  4. Research companies that fit

    Use web research to find startups and important companies that strongly match the clarified thesis. Prefer companies that share the same buyer, workflow, use case, and market boundary. Avoid padding the list with loosely adjacent companies.

  5. Design the proposed landscape

    Create a suggested landscape title, concise description, and one or more categories that reflect the thesis. Assign each company to the best category and include a short inclusion_reason.

  6. Propose before writing

    Show the clarified thesis, proposed title, description, categories, and company list. Ask the user to confirm, refine the thesis, remove companies, add companies, or adjust categories.

  7. Create the landscape after confirmation

    After the user agrees, call create_landscape with the approved title, concise description, and thesis. Create the approved categories with create_landscape_category. Add each approved company URL with add_company_to_landscape and include a short inclusion_reason.

  8. Set expectations after creation

    Tell the user that Axomap has started enrichment for added companies. Do not wait for every background agent to finish in one response. Use get_landscape_enrichment_status later if the user asks whether enrichment is complete.