Convert many file formats, local or remote, to Markdown for LLMs using MarkItDown.
File Systems
Filesystem servers expose files and directories to an AI assistant. Compare allowed-directory controls, search capabilities and write operations. Keep access limited to the folders your task requires.

Direct local file system access.
Remote storage access: SFTP, S3, FTP, SMB, NFS, WebDAV, Git, FTPS, Google Cloud, Azure Blob, SharePoint and more.
Local file system access.
Windows file search using the Everything SDK.
Share code context with LLMs via MCP or the clipboard.
Read from Google Drive and edit Google Sheets.
Directory visualization with semantic analysis, compressed AI-oriented output formats and file categorization.
Filesystem access for browsing and editing files.
Filesystem operations with large-file support: read and write files, sequential reads, directory operations, search and streaming writes with backup.
Edit existing Word .docx files with formatting preservation, tracked changes, comments, footnotes and document comparison.
Read-only full-disk search on Windows over file names, document contents and OCR'd screenshot text.
Access any storage backend through Apache OpenDAL.
Manage macOS Homebrew packages with natural language, get suggestions and troubleshoot brew issues.
Merge files to work within AI chat length limits.
Token-efficient file access: unified-diff edits, targeted grep/sed reads of large files and kernel-enforced confinement to a configured root.
Filesystem access for legacy non-UTF-8 files: detects encodings like CP1251, KOI8-R and GBK, preserves them on write, and handles BOM and line endings.
Read-only filesystem exploration, search and analysis with symlink protection.
Ephemeral file transfer for agents and CI/CD pipelines: upload a file and get a SHA-256-verified, expiring link, then list or delete uploads.
Large-file handling with chunking, navigation, streaming, regex search and LRU caching.
Coherence guard for shared files: when two agents write the same file, the stale writer is denied instead of overwriting, then reacquires and retries.
Read, search, OCR and extract from PDF, TXT, Markdown, DOCX, CSV and image files, with caching and directory confinement.
Box integration for listing, reading and searching files.
Create, read and round-trip Excel workbooks (.xlsx/.xlsm) via a type-safe F# DSL, including regenerating the F# source that reproduces a file.
Convert Markdown files to PDF with customizable themes, headers, footers and styling.
File conversion across image, video, audio, document, data, font, ebook and archive formats, returning a signed download URL.
Access Box content: search, ask questions about files and folders, and extract data.
PDF toolkit: create and analyze PDFs, extract text for RAG, convert to Markdown, split and merge pages, add annotations and form fields, and encrypt files.
Read-only directory listing web app: browse directories, read files, search content, inspect archives, compute hashes and check disk usage.
Local filesystem management with process control, interactive sessions, async search and safe delete via the Recycle Bin.
Read and edit Markdown files by structure (headings, lists, sections) instead of regex or line numbers: read a section, write one list item, append under a heading.
Local document and image tools: read and convert PDF, DOCX, HWP and EML, edit PDFs, resize images, strip EXIF, remove backgrounds and make QR codes.
Share Markdown files and text as readable web links.
Convert, compress, merge, split and OCR PDFs, and convert between Word, Excel, PowerPoint, image, ebook and video formats via the GuruPDF API.
File system operations: read, write, search, copy and move, with path restrictions.
Convert local PDFs into source-linked, card-based standalone HTML readers with notes, highlights and static publishing.
Rename files with descriptive names via Distill, which analyzes PDF, OCR, Office, email and media content, with folder monitoring.
Hash-anchored file edits: read files with line anchors, validate anchors against current content and reject stale writes.
Token-efficient text-file search, reading and refactoring that detects each file's BOM and codepage and preserves them on save.
Give a private ChatGPT app access to selected local folders through OpenAI Secure MCP Tunnel, with per-folder permissions, hash-checked edits and recovery. No shell or public port.
Convert HTML or Markdown to styled PDFs, or render a public URL to PDF, and get a download link.
Google Drive API: search, organize, upload, download, export, share and comment on Drive files.
Read and verify a local media-archive catalog: 3-2-1 backup coverage across disk, NAS, S3-compatible cloud and LTO tape, with read-back hash verification.
Read, convert, edit and compare PDFs on your computer: Markdown with tables, charts and OCR, export to Word, PowerPoint and EPUB, tables to CSV or Excel, merge, split, fill forms and remove metadata. Nothing is uploaded.
Document extraction tooling: read PDFs, text, CSV and JSON, inspect metadata, chunk long documents, validate against JSON Schema and write JSON/CSV output.
Convert PDFs to and from Word, Excel, PowerPoint, JPG, HTML, Markdown and EPUB, plus merge, split, compress, sign and protect PDFs via Convertica.
Yandex Disk and Yandex Wiki integration with unified search, read-only by default, plus granular permissions, path allowlisting and audit logging.
Convert Office documents, images and e-books on your computer to PDF, convert PDFs to Word or images, and merge, split, compress, protect or number PDFs via the conv2pdf API, hosted in France.