Skip to content
GuidesReal EstateMarketingFamilyLive EventseSports

Capability-first field guide

Use Claude for the brief, the critique, and the hard reading.

Claude is useful for long-document analysis, structured briefs, visual inspection, and a separate QA pass. Anthropic says it does not natively output ordinary generated images.

By Ryan WannerPractical guideReviewed August 20, 2026

The analyst-and-reviewer setup.

  1. 01

    Define the reading job

    Name the question, source set, decision, and required structured output.

  2. 02

    Load sources deliberately

    Use clear filenames, document purpose, image order, and any source-of-truth hierarchy.

  3. 03

    Request traceability

    Ask for page/section references, observed image regions, uncertainty, and unsupported-item flags.

  4. 04

    Build the execution brief

    Translate analysis into a prompt, checklist, spec, or handoff for the tool that performs the work.

  5. 05

    Return for critique

    Compare the produced artifact to the original source and brief; focus revisions on material mismatches.

  6. 06

    Save the boundary

    Record what Claude analyzed, what another tool created, who approved it, and what proves completion.

Make analysis and execution two visible steps.

Claude analyst workflow

Design a Claude workflow for analysis, briefing, critique, or visual QA.

JOB
[result]

SOURCE DOCUMENTS / IMAGES
[list]

REQUIRED OUTPUT
[brief, comparison, checklist, rubric, code, etc.]

EXTERNAL TOOL REQUIRED
[image generator, execution environment, database, or NONE]

RISK + REVIEW
[list]

Return: source-loading plan, reusable instruction, document/image analysis steps, structured output, source quote/location strategy, external-tool handoff, second-pass critique, human gate, and completion receipt. State plainly that Claude does not natively output ordinary generated images. Do not describe analysis as execution.

Situation

Claude-assisted virtual staging

Claude receives the original room, writes a property-lock brief, and another supported image generator renders the edit.

Useful output

Claude compares original and staged images for walls, windows, doors, floor, built-ins, camera angle, scale, shadows, and disclosure readiness.

Completion receipt

Original, Claude brief, named render tool, render prompt/output, Claude comparison, human QA, disclosure, and final export.

Keep building the system.

Claude virtual-staging QA

Use the complete paired-tool staging playbook.

Open standalone guide →

Evaluation rubric

Turn critique into observable criteria.

Open standalone guide →

Foundation-model hub

Route render, research, drafting, and critique by job.

Open standalone guide →

Research sources

Check the current rules, capabilities, and source basis.

  1. 01Anthropic: Can Claude produce images? ↗

    Anthropic's current statement that Claude accepts images but does not produce ordinary generated images.

  2. 02Anthropic: Vision ↗

    First-party guidance for providing images to Claude for analysis and visual reasoning.

  3. 03NIST: Generative AI Profile ↗

    Model-neutral guidance for generative-AI risk, testing, documentation, transparency, and content provenance.

Product access, limits, local listing rules, platform policies, and laws can change. Re-check the linked authority before important or public work.

Keep building

Check us out on Skool for 30+ classes and growing.

AI Acceleration: Real Estate is a private, self-directed curriculum for practical AI work across content, communication, lead follow-up, CRM, client research, operations, and relationship management. Build it. Review it. Keep what works.

Explore the Skool community ↗

Follow Ryan + AI Acceleration

No ranking, revenue, closing, or time-saving promise. The classes teach the work and the review systems around it.