The structure of every funder application, on the page where you're applying.
Funding Application Decoder is a side panel that activates on the application portals of the funders you're targeting — NIH F31, Damon Runyon, Helen Hay Whitney, HHMI Hanna Gray, NSF GRFP, Wellcome, MSCA, EMBO, Pew and more. It shows you eligibility against your profile, the required sections with current word and page limits, evaluation criteria with the sub-questions reviewers actually consider, sourced cultural conventions, and content mapping from a local library of past sections you've written.
Your application content is stored in your browser. No AI reads it; nothing is uploaded.
Funder application pages are 30 minutes of scattered reading
Eligibility rules are buried in three places, section limits are in a separate instruction PDF, evaluation criteria are spread across the reviewer guide, and cultural conventions — the stuff that actually distinguishes successful applications — are documented nowhere except in old training PowerPoints and the heads of people who've reviewed before.
- Eligibility surprises kill weeks. Three weeks into a Helen Hay Whitney pre-application before realising the time-from-PhD window doesn't quite work.
- You re-read the same instructions for every cycle. NIH F31 sections, NSF GRFP broader impacts, MSCA's strict 50/30/20 grid — each takes a fresh read because nothing remembers what you learned last time.
- Past sections are stuck in past application folders. Your Specific Aims from one fellowship don't surface when you start another, even when they'd map.
A curated structure, not an AI coach
How it works
Set up your profile once
A 12-question onboarding form — citizenship, career stage, PhD status, time from PhD, institution country, field, prior awards. Stored on your device. Drives every eligibility check.
Visit a funder application portal
The side panel activates with a five-section view: eligibility for you, required sections with limits, evaluation criteria, sourced cultural conventions, and content mapping from your library.
Save sections as you write
Personal Statement, Specific Aims, Sponsor Statement — each saved becomes a library piece. On the next funder you visit, content mapping shows which pieces fit which slots.
What's inside
Per-criterion eligibility verdicts
Each rule on each funder is checked against your profile — pass, pass-with-note, inconclusive, fail-non-blocking, or fail-blocking. Overall verdict is the worst of all rules, with the standard "final determination is by the funder" disclaimer.
Required sections with current limits
Word and page limits per section, format requirements, which sections count toward the page total, sub-section structure where it applies. With a last-verified date and a "may be outdated" banner if it's over 90 days.
Evaluation criteria with sub-questions
The criteria reviewers actually use, with the specific sub-questions they consider for each one — adapted from each funder's published reviewer guide where it exists.
Sourced cultural conventions
Every insight has a typed source — official funder guide, research-administration literature, successful-applicant narrative, or editorial expertise. No AI inference, no unsourced tips, no funder-endorsed claims unless actually endorsed.
A local library that compounds
Section-typed pieces (Personal Statement, Specific Aims, Sponsor Statement, etc.) mapped to the current funder's section requirements with an overlap estimate. Adapt in a side-by-side workspace with word-level diff.
Local deadline reminders
Optional 14-day and 3-day notifications for applications you've added to the tracker — scheduled via chrome.alarms and fired via chrome.notifications. All on-device.
Privacy & Trust
- Your application content never leaves your machine. Profile, library, tracker entries, settings — all in
chrome.storage.localand IndexedDB on your device. - No content scripts. The extension does not read, parse, or modify the content of funder portal pages. It matches the URL against the bundled funder database to decide which structured record to display.
- No
<all_urls>, no "tabs" permission. The host permissions enumerate only the 22 funder domains in the database plus the GradSummit sign-up Worker. On non-funder pages the active tab URL is not even visible to the extension. - No AI runtime, no telemetry, no analytics, no account. The funder database is bundled inside the package; nothing executable is fetched at runtime.
- One optional network call, only if you ask for it. If you tick the box and click Subscribe for product updates, only your email address is sent to GradSummit's sign-up service — never any profile, library, or application content.
Eligibility verdicts are system-generated and the final eligibility determination is always by the funder. Read the full privacy policy.
Built by Dr. Rafiq Muhammad, PhD — author of the Mastering Research book series.
Pricing
Every feature, for everyone — no subscription, no tiers, no account.
All funder records, all five side-panel sections, content mapping, library, application tracker, deadline reminders, and JSON export — all unlocked.
Email is entirely optional and opt-in — leave it blank forever, or add it to hear about occasional updates.
Frequently asked questions
Which funders are supported?
The v1 bundled database covers 10 records: NIH F31 (Predoctoral) and F32 (Postdoctoral); Damon Runyon Postdoctoral; Helen Hay Whitney; HHMI Hanna H. Gray Fellows; NSF GRFP; Wellcome Early-Career Awards; Marie Skłodowska-Curie Postdoctoral; EMBO Postdoctoral; and Pew Biomedical Scholars. The host permissions cover a wider set of domains (e.g. JCC Fund, BWF, CRI, AHA, ASBMB, SfN, CIHR, Searle, Whitehall) so additional records can be added in updates without redeploying the manifest.
Does an AI read or write my applications?
No — and deliberately so. There is no AI runtime anywhere in this tool. It does not draft your aims, rewrite your personal statement, suggest language, or critique your reasoning. You write the application; the tool surfaces the structure and the sourced conventions.
Where do the cultural conventions come from?
Each convention has a typed source field, visible to you on every card. The four source types are: OFFICIAL (the funder's published reviewer guide), LITERATURE (research-administration literature), APPLICANT (published successful-applicant narratives), and EDITORIAL (editorial expertise from GradSummit). Editor-expertise items are labelled as such — they do not pretend to be funder-endorsed.
Can I trust the eligibility verdicts?
Treat them as a system-generated first pass. Every verdict carries the disclaimer "system-generated assessment; final eligibility determination is by the funder," and every funder record carries a last-verified date. Records older than 90 days show a "may be outdated — verify against the funder's current site" banner. The disclaimer is a permanent part of the UI.
Is my application content uploaded anywhere?
No. Profile, library pieces, tracker entries, settings — all stored in your browser's chrome.storage.local and IndexedDB. The only thing the extension can send to GradSummit is your email address, and only if you explicitly opt in to product updates and click Subscribe. No application content is ever transmitted. See the full privacy policy.
How much does it cost?
It's free, with no subscription and no paid tiers. Every feature is available to everyone. There's nothing to buy and no account to create.
Can I get my data out?
Yes. Settings → Export downloads a JSON file containing your profile, library, and tracker — saved to your Downloads folder under your control. Settings → Import accepts a previously exported JSON file to restore on another device. Settings → Delete everything clears all data on your device with one click.
What if I find a bug?
Email support@gradsummit.com — replies go to a real human (the developer).