More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Fillable PDF Studio

    1

    Beautiful fillable PDFs with editable Word sources for work, clients, study and everyday life. Complete free Skill.

    Free

    1 useSecurity scanned

    About this skill

    Good forms. Better follow-through.

    Turn a real task into a polished English form, planner, checklist or practical workpack. Complete native PDF fields, keep the result, and revise the editable Word source when your needs change. Fillable PDF Studio is a complete free Skill with no paid upgrade or required seller service.

    Forty-eight complete scenarios

    Eight families cover work and meetings; freelance and client work; small business operations; learning and teaching; career and professional life; personal and household planning; travel and events; community and interests.

    Create meeting briefs and decision records, client intake and creative briefs, work orders and stocktakes, study weeks and reading journals, interview preparation and onboarding plans, household inventories and moving planners, trip plans and event run sheets, volunteer signup and workshop organisers.

    The creator reference collection contains 48 complete fictional scenarios. Each has a blank native PDF and Word source in US Letter and A4, plus a filled Letter example for acceptance and presentation. All 48 scenarios appear across the six gallery boards. The images show actual generated documents, including native Word sources; they are not AI-generated pictures of a form.

    Sixteen design foundations, shaped around the task

    Bureau, Gridline, Clay, Drafting, Fieldnote, Ledger, Campus, Confetti, Lagoon, Lilac, Signal, Botanic, Waypoint, Gallery, Daybook and Workshop offer editorial margin labels, paired opening sections, response cards, guided rails, precise records and restrained stationery treatments.

    These foundations share native components and are starting points for adaptation. Change the hierarchy, typography, response areas and flow, or create a new composition. A useful one-page checklist stays one page; a longer task gets purposeful continuation pages. The agent chooses the layout from the actual content, rather than forcing every user into a fixed two-page template.

    Truly usable, genuinely editable

    The PDF uses real AcroForm text, multiline, checkbox and dropdown fields where the selected scenario calls for them. Fields have meaningful names, print appearances and matching source definitions. The working copy remains interactive. No JavaScript, macros, automatic form submission or paid design service is added.

    The Word source uses real paragraphs, tables, repeating footer fields and named plain-text response controls. It is intentionally simpler than the more decorative PDF and is not pixel-identical. Word checkbox responses are editable [ ] / [x] text controls; dropdown responses are editable text, not native Word dropdown widgets. A normal Word PDF export is not guaranteed to retain interactive fields. After source edits, reconcile the changed labels, layout and answers, then regenerate the native PDF. This is not automatic two-way synchronization.

    A natural workflow

    Tell your agent who the workpack is for, what needs recording and any paper or brand preferences. Supply only the facts and answers you approve. The agent selects and adapts a relevant scenario, builds the native files, checks the source and renders the pages before delivery. Users do not need to prepare JSON or edit scripts.

    Blank production leaves unknown answers empty and checkboxes unchecked. Fictional examples never become real users' facts. Approved text is not silently shortened to fit. Dates, dropdown choices, zero values and completion checks are validated. If an answer needs more space, enlarge or repaginate the form rather than truncate it. Existing outputs are preserved.

    The intended generated user delivery is a fillable PDF plus an editable DOCX, with brief completion and revision instructions. US Letter and A4 use separate physical geometry rather than scaling a page image.

    Full fictional creator-local example

    Input: Create a two-page client intake for Harbor Paper Co using the Campus foundation, US Letter and English labels. Approved fictional demonstration answers: contact Zoë Miller; email zoe@example.com; project Autumn product launch; service Brand design; target date 2026-11-12. The goal is a clear launch identity and reusable campaign assets. The audience is independent retailers; product photographs and approved descriptions are available. Brand assets and Reference examples are checked; Product information and Approved imagery remain unchecked. The printer needs approved artwork by November 5. Preferred next conversation: 2026-10-23; focus: Confirm deliverables and dependencies. Return a fillable PDF and editable Word source, preserving all answers.

    Actual creator-local output: A two-page native PDF and matching DOCX with 15 response fields/controls. Page 1 contains Contact details (four fields), Project basics (service dropdown and target date) and a multiline goal. Page 2 contains audience/material context, four independent material checkboxes, constraints and the next conversation's date/focus. Canonical PDF values and Word response tags reconciled with the approved source. Both actual documents rendered to two pages. This demonstration was generated and inspected locally, not through an Agensi buyer session.

    Verified locally, with explicit limits

    All 144 blank/filled document pairs passed native field-tree, widget, appearance, bounds and PDF/Word response agreement checks. The 96 blank PDFs completed two programmatic fill/save/reopen cycles, including cleared answers, unchecked boxes, changed dropdowns/dates and a legitimate zero. Named response extraction reconciled for both formats across all 144 pairs. All 288 documents rendered to 576 pages, with every case included in page-overview review and selected pages inspected at larger scale. Custom one-page A4/ink-saver and three-page workpacks also passed native checks from in-memory code and selected recipe dictionaries.

    One real macOS Preview test preserved visible text, dates, multiline notes and checks after save/close/reopen, but Preview rewrote checked-box values and appearance structure. Its saved copy failed strict native checkbox consistency. The Skill detects this, extracts named responses and can regenerate a fresh pair after reconciliation; the recovered 15-field pair preserved responses exactly and passed native checks. Preview dropdown UI selection was not verified. This is a recovery path, not universal reader certification.

    Microsoft Word desktop editing/save/reopen, Adobe Acrobat and other reader UI round trips, keyboard accessibility and PDF/UA certification remain unverified. The supplied helper's verified character scope is English and common Western Latin CP1252; unsupported scripts are rejected instead of corrupted. Other scripts need a separately verified native font workflow. Font binaries and office applications are not bundled.

    Runtime and platform delivery

    A capable agent with Python, file creation and PDF/Word libraries is required. Creator testing used ReportLab 4.4.9, pypdf 6.10.0 and python-docx 1.2.0. Your agent plan, editor and any environment costs remain separate from the free Skill.

    Agensi delivers Skill instructions through its agent connection. This listing does not promise downloaded Skill source, a downloadable template archive or offline Skill access. The published text package contains the instructions, standalone helpers, 48 JSON recipes, design definitions and verification guidance. Load only relevant text into the interpreter; do not save Skill code/assets to disk. Requested PDFs and DOCX files are intended generated user work products.

    Buyer-side multi-file loading, helper execution, dependency availability and generated-file retrieval have not been verified on Agensi. Platform compatibility review is requested before approval. Loading SKILL.md alone does not prove the full workflow. If the host lacks a required file, interpreter, renderer or output capability, the agent must disclose the limitation and must not claim an export that did not happen.

    Frequently asked questions

    Is the complete Skill free? Yes. There is no paid tier, review gate or required seller service. Your agent and editor costs are separate.

    How many scenarios and styles are included? There are 48 complete fictional scenarios in eight families and 16 design foundations sharing native components. All 48 appear in the gallery. The agent can adapt them or compose a new workpack.

    Are PDF and Word both editable? The PDF has native interactive fields. The Word source has native text, tables and tagged plain-text response controls. Word checkbox/dropdown sources are editable text, and manual changes do not automatically sync back to the PDF.

    Do I always get two pages? No. The examples use two purposeful pages, but production page count and response space follow the actual task. The helper supports 1–20 purposeful pages and refuses overflowing content rather than dropping it.

    Can I reuse a completed file? Named answers can be read from a generated PDF or Word source without changing it. Review static edits and extraction notes, then regenerate a new native pair. A changed field structure requires explicit reconciliation.

    Does every reader preserve the fields? No. Test a saved copy in your intended reader. The observed Preview checkbox issue is detected and has a verified regeneration path; other reader UI round trips and Word desktop remain unverified.

    Does Agensi guarantee all supporting files and document downloads? The creator-local workflow is verified, but Agensi buyer-side multi-file loading, execution and generated-file retrieval are unverified. The listing requests platform review and does not promise downloadable Skill source or offline access.

    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 any payment.

    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

    1 use

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 8 days ago

    • Free with an Agensi account

    Requires

    Python

    Listed8 days ago

    What's inside

    Frequently Asked Questions

    Free