Feedvault

Plain-language documentation

Bring your own inputs.
Keep control of the result.

These tools run in the page you open. They have no server-side processing, API key, remote extraction service, or automatic connection to your accounts.

Download the JSON catalog · Read the plain-text guide

Privacy and storage

Your pasted text and chosen files are processed in browser memory. The tools do not upload them, put them in page URLs, or send them to analytics. Tool pages do not load Microsoft Clarity, affiliate tracking, or remote fonts. Normal requests to load the website still reach its hosting provider.

The timestamp notebook offers optional on-device saving. It starts off. Enabling it saves that notebook in this browser’s localStorage, which is not a backup or cross-device sync. Its clear action removes the saved notebook. Other tools use page memory only. Use the export buttons before you leave, and use Clear to remove working input and results. Your browser may independently restore form values, manage downloads, or sync its own data.

The optional notebook video player is not loaded until you click its load button. That action connects to YouTube’s privacy-enhanced embed service, which receives ordinary connection information and handles playback under YouTube’s policies. Manual note-taking works without it.

If you choose an AI assistant to provide text or read results, the data you share with that assistant is subject to its own processing and privacy practices. Browser-local computation is not a promise that an assistant keeps your conversation on this device.

Input limits and safe exports

For browser assistants

Use the normal browser interface: open the specific tool, enter only material the user has authorized, activate the labeled action, and read the visible output. Downloads are optional. Read-only search fetchers and web crawlers can read the documentation but do not execute these tools.

The catalog is a custom documentation format with a version, tools, accepted input and output formats, examples, limits, and privacy details. It is not a standardized registration manifest and does not expose a callable HTTP endpoint.

Optional local transcript site tool

The transcript cleaner progressively detects document.modelContext.registerTool and registers feedvault_clean_transcript where supported. It uses the same local transformation as the visible tool. Its arguments are text (required string), removeTimestamps (optional boolean, defaults to true), and joinLines (optional boolean, defaults to true). Input is limited to 2 million characters. It returns JSON with ok, output, and warnings, or an error code and message.

The registered function reads only those arguments. It does not read your current form, selected files, notebook, clipboard, cookies, or browser storage; it does not write results to your page or save them. Availability depends on your browser, account, and chosen assistant. The standard is evolving, and the normal interface remains available when it is unsupported. See Chrome’s WebMCP documentation and OpenAI’s site-tool guide.

Example request: Open the transcript cleaner and clean only this SRT text: “1 [newline] 00:00:00,000 --> 00:00:02,000 [newline] Welcome to the lesson.” Remove timestamps and join lines. The expected cleaned text is “Welcome to the lesson.”

Tool-by-tool reference

YouTube timestamp notebook

Turn a lecture, tutorial, or interview into a small set of moments you can find again. Add your own timestamps and notes, then take the notebook with you.

Accepts: YouTube URL (optional), Video title, Manual timestamps and notes. Produces: Linked Markdown notebook, Plain-text notes, Structured JSON.

  1. Add a video URL and a title you will recognize later. The title is yours to enter; this tool does not look it up.
  2. Watch in a separate tab, or explicitly load the optional YouTube player. Pause at a useful moment and type its timestamp and your note.
  3. Use Edit to revise a saved note or Remove to delete it. The list follows your entry order so you control the outline.
  4. Download Markdown for a linked notebook, TXT for plain text, or JSON for structured data. Enable device saving only if you want a local copy.

Example: For a lecture on feedback loops, add “0:45 — Define the system boundary” and “3:20 — What changes when the delay increases?” A timestamp preserves the context; a question gives you something to recall later.

Transcript cleaner for SRT, VTT, and pasted text

Turn a transcript you already have into something you can read, quote, or study. Remove caption formatting, choose whether to keep timestamps, and preview the result before applying a change. Your text stays in this page.

Accepts: Pasted transcript text, Local SRT, VTT, or TXT file, Formatting choices. Produces: Cleaned TXT, Normalized SRT or VTT with original cue times, Preview and cleanup warnings.

  1. Paste your transcript or choose a local UTF-8 SRT, VTT, or TXT file. Automatic detection recognizes subtitle timing arrows and the WEBVTT header; choose TXT explicitly for literal timing-like prose.
  2. Choose whether to join wrapped lines, remove recognized formatting tags, or remove whole non-speech lines such as [Music]. Keep timestamps if you need a reference back to the recording.
  3. Preview the cleaned text. Download TXT for reading, or SRT/VTT when the input provides complete cue timings. Use the TXT preview as a new input only if you want to edit it further; Undo restores the previous source.

Example: The example has three cues starting at 00:00:01.250, 00:00:05.000, and 00:00:09.125. Joining lines turns “Learning begins with” and “a good question.” into one sentence. Removing [Music] changes caption text without moving any surviving subtitle start or end time.

YouTube timestamp link & chapter builder

Make the useful parts of a video easy to share. Paste timestamp labels once to build direct links and check whether your outline meets the basic chapter-format rules.

Accepts: YouTube video URL, Timestamp and label lines, Optional video duration. Produces: Direct timestamp links, Chapter TXT, CSV link table.

  1. Paste a YouTube video URL. Standard watch, share, Shorts, and embed URLs are accepted; unrelated websites are rejected.
  2. Enter one timestamp and label per line, for example “1:30 Worked example”. Keep the intended chapter order.
  3. Add the total video duration if you want to check the final chapter length. Generate links and read any chapter warnings.
  4. Download the chapter TXT for a video description, or CSV for a spreadsheet with one direct link per moment. You must add chapters to your own video in YouTube Studio yourself.

Example: A 4:00 tutorial could use “0:00 Overview”, “0:45 Setup”, and “2:10 Worked example”. Those starts create chapters of 45, 85, and 110 seconds. The same entries also become individual links you can share in lesson notes.

Playback Speed Calculator & Video Study Planner

A video list is not a study plan. Budget the time to watch, then leave room to work through what you learned. This planner fits both into the days you actually have.

Accepts: Video titles and durations, Playback speed and practice allowance, Weekdays, session length, and deadline. Produces: Watch and practice time totals, Dated session plan and feasibility warning, Checklist TXT and all-day ICS.

  1. Enter one video per line as Title | MM:SS, or use HH:MM:SS for a longer video. Keep the list in the order you want to study.
  2. Choose playback speed, session length, and the extra practice minutes you want after each video. Practice time is separate from viewing time.
  3. Select available weekdays and an inclusive start-to-deadline window. Build the plan to see how many sessions fit and whether work remains.
  4. Download a plain-text checklist or an ICS calendar. Review the all-day reminders before importing; choose your own time of day in your calendar.

Example: Three videos lasting 20, 30, and 40 minutes total 90 minutes of footage. At 1.5× they need 60 minutes of watching. Add 10 practice minutes per video for 90 minutes of study: three 30-minute sessions. A Monday-to-Friday week with Monday, Wednesday, and Friday selected fits the plan exactly.

YouTube subscription organizer and OPML exporter

Turn a subscription export into a list you can actually work with. Combine supplied CSV or OPML files, sort channels into your own groups, and keep a portable copy on your device.

Accepts: YouTube subscriptions CSV, OPML outlines, Manual channel IDs, URLs, or handles. Produces: Grouped, deduplicated channel list, CSV download, OPML download.

  1. Export the YouTube data you need with Google Takeout, then extract your subscriptions CSV on your own device. You can also use an existing OPML file.
  2. Select your files or paste the contents. CSV needs Channel Id or Channel Url; Channel Title and Groups are optional. Use semicolons between groups.
  3. Choose Import and merge. Review skipped entries and duplicate counts. Add individual channels manually, edit groups in the table, or remove channels from this local list.
  4. Export CSV for a spreadsheet or OPML for compatible readers. The group edits are saved in the download; the tab itself does not persist your list.

Example: Import the example to see Science and Mathematics groups. A channel ID and a handle that might represent the same channel stay separate, because verifying that relationship would require a live lookup. Importing the same channel ID again combines its groups.

YouTube Citation Generator: APA, MLA & Chicago

Keep your sources attached to your ideas. Enter the details you can verify, then create a citation draft with clear formatting cues and no guessed metadata.

Accepts: Video title and URL, Uploader and optional creator, Publication details and optional timestamp. Produces: APA 7, MLA 9, or Chicago 18 reference, Parenthetical citation or full footnote, Clipboard copy and TXT export.

  1. Choose APA 7, MLA 9, or Chicago 18 notes and bibliography. Paste a YouTube video URL and enter its title manually.
  2. Enter the uploading account name. For MLA or Chicago, add the identifiable creator; choose whether that name is a person or a group. APA treats the uploader as author.
  3. Add the upload date when known. You may also add your actual access date, a quotation timestamp, and the video’s duration. The tool never substitutes today for a missing date.
  4. Review the reference and its in-text citation or footnote. Copy with formatting into a compatible editor, or download plain text. Check italics, hanging indents, and your assignment requirements.

Example: The example uses the APA publisher’s sample video by Above The Noise, dated 18 October 2017. APA places the uploading account first and italicizes the video title. Switching to MLA changes the container and date format; Chicago returns a bibliography entry and a full footnote.

Free Cornell Notes Template & PDF Generator

Give your notes a job after the video ends. Build a Cornell-style sheet with space for the main ideas, questions that test your recall, and a short summary in your own words.

Accepts: Topic and source, Cues, notes, and summary, A4 or Letter; filled or blank. Produces: Paginated A4 or Letter PDF, Printable Cornell-style sheet, Plain-text notes.

  1. Name the topic and add a source, such as a video title, URL, chapter, or lecture date.
  2. Record the main ideas in Notes. Put questions, terms, or timestamp cues in the narrower left column.
  3. Write a brief summary after the session. To review, cover the Notes column and answer the cues from memory.
  4. Choose A4 or Letter and preview your sheet. Download a paginated PDF, print a filled or blank template, or export TXT. PDF creation happens on your device.

Example: Topic: Feedback loops. Cue: “What makes feedback negative?” Notes: “A response counteracts a change in the measured quantity; negative does not mean harmful.” Summary: “Feedback describes a relationship, while delay affects how the system responds.”

Search multiple transcripts for a word or phrase

Find the passage you remember across transcripts you supply. Search a literal word or phrase, compare its surrounding context, and carry useful matches into research notes with a CSV or text export.

Accepts: Multiple local SRT, VTT, or TXT files, Literal search phrase, Optional per-file source links. Produces: Matching passages with surrounding context, Cue timestamps when supplied, CSV and TXT search exports.

  1. Select several local transcript files or add pasted text to the collection. Each file is validated before it is added, so a broken subtitle block cannot quietly hide a result.
  2. Optionally enter a source URL beside each transcript. For recognized YouTube video links, timed matches open at the cue’s start second. Other HTTP or HTTPS links remain the source link you supplied.
  3. Enter a word or phrase, choose case matching and the amount of context, then search. Browse 50 passages at a time or download every match with filenames, timestamps, context, and source links.

Example: Load the two-file example and search for “question.” The SRT contributes three timestamped passages, and the TXT notes contribute two untimed lines. Search for “a good question” to match a phrase split across wrapped lines inside one subtitle cue.

YouTube playlist snapshot comparison and dedupe

See what changed between two lists you already have. Paste video URLs or import snapshots to compare additions, removals, shared videos, and their relative order, entirely in your browser.

Accepts: Two supplied URL/ID lists, CSV snapshots, Bookmark HTML exports. Produces: Added, removed, common, and reordered videos, Deduplicated comparison, CSV or TXT result download.

  1. Put the earlier snapshot on the left and the later snapshot on the right. Paste one video URL or 11-character video ID per line, or choose TXT, CSV, or exported bookmark HTML.
  2. For CSV, include a Video Id or URL column and optionally a Title column. A selected file replaces pasted input on that side. Leave a side blank to compare against an empty list.
  3. Choose Compare snapshots. Review duplicate and skip counts for each side. The first occurrence of each video ID sets its input position.
  4. Filter to additions, removals, common videos, or reordered videos. Export every result in that filter as CSV or TXT, including results beyond the visible page.

Example: The example contains one repeated video in the earlier snapshot. The later snapshot adds one video, removes one, and reverses the relative order of the two shared videos. Adding a new video before an otherwise unchanged shared sequence would not count as reordering.

Anki flashcard exporter for your own study notes

Make a small, deliberate set of Anki notes from material you choose. Write a question and answer, or mark a cloze deletion yourself. Preview, edit, and export notes without connecting an Anki account.

Accepts: Manually written Basic or Cloze notes, Optional selected source text, Optional source URL. Produces: Editable note collection and plain-text preview, Separate Basic and Cloze TSV files, Anki import instructions.

  1. Write a note directly, or paste or import text you are allowed to use. Select a passage to copy it into a Basic answer or a Cloze text field.
  2. For Basic, write both a question and an answer. For Cloze, select the words to hide and choose a deletion number, or type a flat marker such as {{c1::retrieval practice}}. Add an optional source link and preview the note.
  3. Add notes to the collection, editing or removing any that need work. Download Basic and Cloze notes separately; one cloze note can produce several Anki cards when it uses different deletion numbers.
  4. In Anki, import the TSV with the matching Basic or Cloze note type, Tab separator, and Allow HTML enabled. Map columns to Front / Back or Text / Back Extra, choose your deck, and check the duplicate handling before importing.

Example: A Basic note asks “What is retrieval practice?” and answers “Recalling what you know without looking at the source.” The Cloze example hides retrieval practice inside a sentence. The downloads keep these note types in separate files so you can choose the correct Anki import mapping.

Search YouTube watch history from Google Takeout

Find a video in a watch-history export without handing over your account. Open only the relevant Takeout JSON or HTML files, search the fields they contain, and download your matching records.

Accepts: Selected Takeout watch-history JSON or HTML files, Title, channel, and UTC date filters. Produces: Searchable history records with supplied source links, Missing-field and skipped-record counts, Matching-record CSV download.

  1. Use Google Takeout to export the YouTube history you want. Download and extract the archive yourself; this tool never needs the full archive or unrelated Google data.
  2. Choose the extracted watch-history.json or watch-history.html files and select Open history files. Importing valid files replaces the current in-tab results.
  3. Check the accepted, skipped, and missing-field counts. Search title or channel text, add a channel filter, or limit results to an inclusive date range in UTC.
  4. Review source links and original timestamps where supplied. Export all matching records as CSV. Clear all to remove the imported data and search terms from the tab.

Example: Load the illustrative example, then search for “curiosity” or filter the dates from September 28 through October 1, 2026. One record has no channel, and another has no video link. Missing information is shown explicitly rather than guessed.

Review calendar generator

Give useful ideas another visit. Choose when to return to a topic, see the dates together, and take the schedule to the calendar or spreadsheet you already use.

Accepts: Topics with optional learning dates, Default start date, User-chosen day offsets. Produces: Chronological review preview, All-day ICS calendar, CSV schedule.

  1. List the topics, lessons, or videos you want to review. Add an optional learning date after a pipe: Topic | YYYY-MM-DD.
  2. Set a default learning date for lines without their own date. Enter the review offsets you want, such as 1, 3, 7, 14.
  3. Build the calendar. Each number is measured from that topic’s learning date, rather than from the previous review. Dates are sorted for a quick workload check.
  4. Download ICS for all-day calendar reminders or CSV to edit in a spreadsheet. Import only after checking your dates and the number of events.

Example: A topic learned on 28 February 2028 with review offsets 1, 3, 7, 14 is scheduled for 29 February, 2 March, 6 March, and 13 March. Leap days are preserved. A second topic can use a different learning date on its own line.

Subtitle vocabulary builder with word counts and context

See which words recur in a transcript you want to study. Start with frequency and the original context, remove words you already know, and add your own meanings before exporting a personal vocabulary list.

Accepts: Subtitle or plain text, Optional known-word list, Your own definition notes. Produces: Word frequency and example contexts, First supplied timestamp per word, CSV and TSV vocabulary lists.

  1. Paste a transcript or choose a local SRT, VTT, or TXT file. Subtitle timecodes are parsed separately from the words, and recognized caption formatting is removed.
  2. Add an optional list of known words and set the minimum word length. Build the list to count the remaining words, ordered by frequency.
  3. Review up to three contexts per word and add a definition or study note in your own words. Browse the list in pages of 50, then download all rows as CSV or TSV.

Example: The sample transcript uses “question” three times and “answer” twice. Adding “question” to your known-word list removes it on the next build. A note such as “a response that you can test” can be attached to “answer” and travels with its contexts in the export.

Review before you rely on it

These are deterministic formatting and planning aids. Review citations against your style guide and source metadata, check calendar dates before import, and keep copies of original exports. The tools do not guarantee a learning outcome, verify video availability, or automatically move anything into the Feedvault app.