More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Structured Thinking — Evidence & Mind Maps

    1

    Turn supplied notes, proposals or research excerpts into a concise explanation of what they support and what remains uncertain.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned

    Structured Thinking — Evidence & Mind Maps

    Example session with this skill installed

    “Across four weeks our fictional event had 80 registrations and 52 attendees. Of 28 people who answered a feedback form, 20 said the time was too late. Should we move the event earlier? Summarize the reasoning in English and generate a mind map.”

    • Read your context and instructions
    • Compiled the structured thinking

    First confirm feasible times and run a small comparison. The supplied information does not show that an earlier slot would improve attendance.

    Reported facts: 52 of 80 registrants attended, a 65% attendance rate. Twenty of 28 feedback respondents said the time was too late. That self-selected feedback must not be treated as a result for every registrant.

    Inference: timing may affect some participants, but these records do not establish causation. Other reasons for missing the event may matter.

    Unknowns: nonrespondents' reasons, feasible time slots, and the cost of a different slot are not supplied.

    Proposed next steps: ask about feasible times and nonattendance reasons, check venue and staffing costs, then compare small sessions using measures agreed in advance.

    The accompanying generated map places “Test feasible times before a full switch” at the center, linked to the attendance figures, the feedback denominator, the causal limitation, and the missing information. The listing screenshot is an actual fictional example, not a promise that every future generation will render text perfectly.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    A clear brief that keeps the evidence, the interpretation and the gaps distinct.

    Turn supplied notes, proposals or research excerpts into a concise explanation of what they support and what remains uncertain. Lead with the main conclusion when one is justified; say what is missing when it is not.

    Use the written brief on its own, or request a matching mind-map image through a compatible image tool. Both are designed to preserve important qualifiers, dates, units and sample sizes instead of making a tidy diagram look more certain than the evidence.

    Your output

    • A concise, conclusion-first brief.
    • Clear labels for source-reported facts, inferences and unknowns.
    • Relevant source attribution and limits in the written explanation.
    • A matching image when requested and supported by your host, with visual review before delivery.
    • Practical next questions when your task calls for them.

    Try it

    “Orders rose from 8 out of 100 visits to 10 out of 200 visits, but the ad channels changed. Does this show that the new page converts better? Give me a text-only brief with the arithmetic and uncertainty.”

    The observed rate changed from 8% to 5%, even though order count increased. A different traffic mix prevents attributing that change to the page alone. This is the kind of distinction the workflow is built to retain.

    Requirements and limits

    The written workflow needs an AI assistant that can read your material and the skill resources. English and Chinese are supported. Images additionally require a real image-generation tool, image inspection and a supported way to deliver the file. Model access, credits, subscriptions and provider fees are not included.

    An optional Python 3.9+ standard-library helper checks a prepared data structure and produces summary text, a copy ledger and an image brief. It does not independently reason, verify sources or generate an image. A prompt is never presented as a finished visual.

    Generated images may need corrections, especially for text. If the requested image cannot be made, you still receive the useful written brief with that limitation stated. Independent browsing occurs only when separately requested and available.

    This is an original reasoning workflow, not a validated assessment, formal proof, authorized book adaptation or substitute for professional medical, legal or financial advice. Destination-host loading and buyer delivery must be verified separately from local development tests.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Research & Analysis.

    Requires host image-generation tools for visual output. Does not verify source truth independently. Limited to English and Chinese languages. Visual QA is only possible if the host supports it.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 11 days ago

    Software
    Image-generation tool

    Needed for the full visual workflow; your agent must also be able to view the generated image.

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Network Access
    Listed11 days ago

    What's inside

    Frequently Asked Questions

    It helps your assistant turn supplied material into a concise brief that separates source-reported facts, inferences and unknowns. For decisions, it can also organize options, counterevidence, important assumptions and a proposed next test.

    No. You can request a text-only brief. A matching image is optional and requires suitable generation, inspection and delivery tools. If a requested image cannot be produced, the assistant should state that clearly rather than present a prompt as a finished visual.

    No. Evidence labels organize the analysis; they do not establish that a source is true or a conclusion is correct. Independent checks require accessible sources and suitable tools. The optional Python helper checks prepared structure and references, not factual accuracy.

    It records options, source-linked claims, counterevidence, the uncertain assumption most likely to change the conclusion, and a bounded next test. Updates can show what changed. A review date or trigger is a planning note, not an automatically scheduled reminder.

    The text workflow needs an assistant that can read the Skill and your material. The optional helper requires Python 3.9+ and the standard library. Image tools, model access, subscriptions and generation credits are separate.

    The helper produces Markdown summary text, a JSON copy ledger, a text image brief and validation JSON. Image formats depend on the chosen generator. Local helper tests do not establish compatibility, loading or file delivery on every assistant platform.

    New in Research & Analysis