AI agents that help you summarize and question documents and PDFs
171 AI agents and agent tools that help you summarize and question documents and PDFs. Ranked by momentum this week, with setup difficulty for each.
Ranks for week 44 (Oct 26 to Nov 1, 2026). Data as of Nov 2, 2026.
In short
- Gaining the most momentum this week: elephant-goldfish-model, Survival LLM, and DeepTutor.
- Easiest to start with (nothing to install, or adds to an AI app you already use): elephant-goldfish-model, PPT Master, and UNIGIT.
- #1
elephant-goldfish-model
Elephant-goldfish-model is a skill that helps teams plan software features with AI agents and stress-test the design before coding.
- #2
Survival LLM
Survival LLM is an agent app that answers emergency and survival questions offline on your own computer, citing its sources, for households and anyone without reliable internet.
- #3
DeepTutor
DeepTutor is a self-hosted AI tutor that turns your own documents and notes into personalized lessons for learners.
- #4
PaperPilot
PaperPilot is an MCP server that lets your AI assistant merge, split, compress, and OCR PDFs on your own machine.
- #5
Hob
Hob is a desktop app that runs open-weight AI models on your own computer for private, offline use, including filling out PDFs and working through documents.
- #6
ex-USSR Kinship Research
ex-USSR Kinship Research is a skill that helps AI agents trace family history in former Soviet records for people researching ancestors from the region.
- #7
Capri
Capri is a Windows desktop agent that carries out file, project, and computer tasks from plain-language requests using Claude Code, Codex, or Antigravity.
- #8
XERJ
XERJ is a local search engine and MCP server that lets AI agents search code, docs and logs without reading files one by one, for developers and coding-agent users.
- #9
PPT Master
PPT Master is a skill that turns documents or topics into editable PowerPoint decks for people who need slides fast.
- #10
UNIGIT
UNIGIT is an AI workbench that brings models, tools, and files into one place so people can get real work done for everyday professional tasks.
- #11
DS Bot
DS Bot is a plugin that gives you a team of long-lived bots that hand work to each other in group chats.
- #12
Agenvoy
Agenvoy is a self-hosted agent that runs on your own computer and completes multi-step work like research and file tasks for people who want results, not just answers.
- #13
graphify
Graphify is a skill that turns a project's code, docs, and PDFs into a queryable map for AI coding assistants, so they can answer questions without rereading every file.
- #14
diagram-design
Diagram Design is a skill that draws editorial diagrams as single HTML files for people using AI coding agents.
- #15
headcount
Headcount is a set of skill plugins that gives Claude Code or ChatGPT specialist departments like finance, security, and marketing for business questions.
- #16
shuohao-skills
shuohao-skills is a set of skills that helps an AI agent produce short-drama videos, from character bibles and outlines to screenplays and storyboards, for creators making AI short dramas.
- #17
text-to-cad
text-to-cad is a plugin that lets an AI agent design physical products as 3D CAD models from text descriptions for makers and engineers.
- #18
Archify
Archify is a skill that turns an idea, plan, or codebase into an interactive diagram for people who need to understand or share a system.
- #19
Helix Foundry
Helix Foundry is an agent app that maps company data into a queryable graph for analysts and business teams.
- #20
vibe-knowledge-film
vibe-knowledge-film is a skill that turns a script or topic into a short explainer video in the style of popular Douyin knowledge clips, for people who want to make such videos without a video editor.
- #21
Answer me with HTML
Answer me with HTML is a skill that answers hard questions with a readable one-page HTML page for anyone.
- #22
career-ops
Career-ops is an open-source job search agent that scores job listings against your CV and drafts tailored applications for you to review and submit.
- #23
OpenMuse
OpenMuse is a self-hosted personal agent that browses the web, runs terminal commands, and handles files for people who want a delegated assistant.
- #24
WeChatBridge
WeChatBridge is a native macOS tool that sends selected WeChat chat history to AI agents or Obsidian for summaries and notes.
- #25
Arsaze
Arsaze is a video editor that lets an AI agent or you edit a real timeline, for video creators and their teams.
- #26
GenOffice
GenOffice is a free, open-source office suite that lets you edit Word, Excel, PowerPoint and PDF files with an AI agent for people who work with documents.
- #27
WinShun
WinShun is a free, open-source Windows search launcher that finds files, text inside documents, and apps by name or pinyin.
- #28
cli-anything-wps
cli-anything-wps is a toolkit that lets AI agents control WPS or Microsoft Office documents, slides, and spreadsheets for people working on Windows.
- #29
Atticus
Atticus is a macOS voice assistant that moves windows, finds files, and runs tasks on your Mac for people who prefer speaking to typing.
- #30
QwenPaw
QwenPaw is a self-hosted personal AI assistant that runs tasks, remembers context, and connects to chat apps like DingTalk, Telegram, and Discord.
- #31
arena-skill
Arena-skill is a Claude Code skill that makes many Claude sub-agents compete in a bracket so you get a stronger answer when Claude's first attempt is poor.
- #32
Tavotto
Tavotto is a desktop app that polishes AI-drawn or script-made scientific figures for journal submission for researchers and academics.
- #33
Humanizer-zh
Humanizer-zh is a Claude Code skill that edits Chinese text to remove stock AI phrasing while keeping facts and meaning for writers and editors.
- #34
ArcReel
ArcReel is a self-hosted agent app that turns novels, scripts, or product material into consistent short videos for video creators.
- #35
Synthadoc
Synthadoc is an open-source tool that turns raw documents into a structured, local Markdown wiki that stays accurate as sources grow, for individuals and teams.
- #36
Understand-Anything
Understand-Anything is a plugin that turns a codebase, knowledge base, or docs into an interactive graph you can explore and ask questions about, for people new to a project.
- #37
drama-skills
Drama Skills is a set of skills that adds AI short drama and micro-drama creation to Claude Code or Codex, from script to storyboard to video prompts, for writers and studios.
- #38
MOVO
MOVO is a self-hosted platform that turns DeepSeek's agent runtime into a team workspace with knowledge search, research, and admin controls for organizations.
- #39
Academic Research Skills
Academic Research Skills is a set of Claude Code skills that guides researchers through research, writing, review, and revision for a paper.
- #40
yomiyasu
yomiyasu is a skill that rewrites AI-generated Japanese into natural, easy-to-read Japanese for Claude Code, Codex, and Cursor users.
- #41
Carrel
Carrel is a self-hosted knowledge base that lets AI agents search your own documents and their related facts for accurate answers.
- #42
Superhero Skills
Superhero Skills is a skill plugin that gives AI agents structured ways to run product discovery and design work for product builders.
- #43
EdgeEver
EdgeEver is a self-hosted notes and knowledge base app that lets AI agents read and write your notes through MCP, for people who want an Evernote or Obsidian alternative.
- #44
Kept
Kept is a self-hosted home inventory that tracks what you own and where it is, with an AI assistant for households.
- #45
MLView
MLView is a skill and VS Code viewer that turns an ML workflow into a source-linked interactive diagram for people reading ML code.
- #46
hyperresearch
Hyperresearch is a deep research skill that turns Claude Code or Codex into a research agent that builds a searchable source library for researchers and analysts.
- #47
book-to-skill
book-to-skill is a skill that turns a technical book PDF into a structured skill your AI agent can search chapter by chapter, for people who want to reuse a book's content while they work.
- #48
MemPalace
MemPalace is a local memory system that stores your AI conversation history and searches it later for people who want their AI tools to remember past work.
- #49
Floe Agent
Floe Agent is an open-source AI agent workspace for iPad, iPhone and Android that keeps notes, files, code and a Linux environment on your device.
- #50
Presenton
Presenton is an open-source app that generates editable PowerPoint and PDF presentations from a prompt or document for business and education users.
-
Open WebUI
Open WebUI is a self-hosted AI interface that runs local and cloud models, with tools and agent features, for people who want their own chat setup.
-
MinerU
MinerU is a document parser that turns PDFs and Office files into clean markdown or JSON for AI workflows.
-
ai-job-search
ai-job-search is a framework that runs on Claude Code to evaluate job postings, tailor CVs, and write cover letters for job seekers.
-
AnythingLLM
AnythingLLM is an app that lets you chat with private documents and run AI agents on your own computer or server.
-
Cherry Studio
Cherry Studio is a desktop app that chats with multiple AI models and runs agents for everyday work on your computer.
-
obsidian-skills
obsidian-skills is a set of skill files that teaches AI coding agents to read and write Obsidian notes, bases, and canvases for note-taking users.
-
nanobot
nanobot is a self-hosted personal AI agent that works with files, remembers preferences, and schedules tasks for people who want their own assistant.
-
CowAgent
CowAgent is an open-source personal AI assistant that plans tasks, runs tools and skills, and keeps memory for people who want a 24/7 agent on their own computer or server.
-
SiYuan
SiYuan is a self-hosted knowledge workspace that lets people and AI agents work on notes together, for people who want private, local-first notes.
-
Reactive Resume
Reactive Resume is a free, open-source resume builder that helps job seekers write, check, and track resumes and cover letters.
-
reverse-skill
reverse-skill is a skill pack that routes AI coding agents to reverse engineering and authorized security testing methods for security researchers.
-
Vane
Vane is a self-hosted AI answering engine that searches the web and gives cited answers for people who want private searches.
-
Next AI Draw.io
Next AI Draw.io is a tool that draws and edits draw.io diagrams from chat instructions for people who need diagrams without drawing them by hand.
-
AionUi
AionUi is a free, open-source desktop app that lets you run AI agents on your computer for file work, coding, and automated tasks.
-
agent-skills
Agent-skills is a collection of skill packs that teach AI coding agents to audit Vercel projects, review UI and writing, and deploy sites for developers.
-
OfficeCLI
OfficeCLI is a command-line tool that lets AI agents read, edit, and create Word, Excel, and PowerPoint files without Office installed.
-
GPT Researcher
GPT Researcher is an agent that runs deep web and local document research and writes cited reports for people who need factual summaries.
-
TencentDB-Agent-Memory
TencentDB-Agent-Memory is a self-hosted memory hub that lets a team's AI agents share saved chats, skills, docs, and code context for developers.
-
FastGPT
FastGPT is a platform that builds AI agents and question-answering systems over your own documents for teams that want a visual setup.
-
Sim
Sim is a workspace that lets you build, deploy, and monitor AI agents and workflows without much code.
-
Distilly
Distilly is a skill creator that turns a person's messages, documents, and public sources into a reusable profile for AI agents.
-
OpenWork
OpenWork is a desktop app that runs AI agents on your own files for people who want a free alternative to Claude Cowork.
-
MaxKB
MaxKB is an open-source platform that builds enterprise agents for customer service and internal knowledge bases.
-
khazix-skills
Khazix Skills is a collection of ready-made skills that teach AI agents like Claude Code and Codex to handle tasks such as disk cleanup, writing, and project notes.
-
Douyin_TikTok_Download_API
Douyin_TikTok_Download_API is a self-hosted app that downloads TikTok and Douyin videos without watermarks and collects their data for you.
-
cli
CLI is a command-line tool that lets people and AI agents work with Lark/Feishu documents, chats, calendars, and mail for business teams.
-
SurfSense
SurfSense is a desktop app that researches, transforms, and edits your private documents on your own machine, as an open-source alternative to NotebookLM.
-
claude-obsidian
claude-obsidian is a set of Claude Code skills that turns sources you save into a linked Obsidian notes wiki for people who want to keep their research organized.
-
Claudian
Claudian is an Obsidian plugin that puts AI coding agents like Claude Code inside your notes vault for working on your notes.
-
Skill Seekers
Skill Seekers is a tool that turns documentation sites, GitHub repos, and PDFs into Claude skills for developers and AI builders.
-
AutoResearchClaw
AutoResearchClaw is an agent app that turns a research idea into a full academic paper with citations and experiments, for researchers and students.
-
NoteGen
NoteGen is a desktop and mobile Markdown note app that uses AI to turn scattered captures into organized notes for personal knowledge work.
-
Kami
Kami is a skill that gives AI agents layout rules and templates to export polished PDF, PNG, and editable PowerPoint documents for writers and business users.
-
cangjie-skill
Cangjie Skill is a skill generator that turns books, long videos, and podcasts into reusable AI skills for applying their methods.
-
InkOS
InkOS is a story creation agent that drafts, reviews, and revises novels, scripts, and translations for writers and content creators.
-
dashi-ppt-skill
dashi-ppt-skill is a skill that builds editable slide decks from your documents for business reports, pitches, and training.
-
Plannotator
Plannotator is a local browser review surface that lets you mark up agent plans, documents, and code diffs and send feedback back to the agent for people who review AI coding work.
-
Local Deep Research
Local Deep Research is an agent app that runs multi-step research across the web, academic papers, and your own documents for people who want privacy and control.
-
html-anything
html-anything is an agent app that turns notes, data, or Markdown into a ready-to-share HTML page for people who want clean web-style output without writing HTML.
-
FileCodeBox
FileCodeBox is a self-hosted file and text sharing tool that lets an AI app create share codes for people to pick up files for them.
-
RikkaHub
RikkaHub is an Android app that lets you chat with AI models from many providers in one place.
-
Oh Story
Oh Story is a skill pack that helps writers plan, draft, analyze, and polish Chinese web novels inside Claude Code, Codex, and similar coding agents.
-
Repowise
Repowise is a codebase index that lets coding agents answer questions about code, callers, risk and tests without re-reading every file.
-
Product-Manager-Skills
Product-Manager-Skills is a library of 77 product management frameworks that teaches Claude, Codex, and similar agents to write PRDs, plan discovery, and set priorities for product managers.
-
ODS
ODS is a self-hosted AI server stack that installs local models, chat, voice, workflows, and document search on your own computer for people who want private AI at home.
-
Yuxi
Yuxi is a self-hosted platform that gives teams a private knowledge-base assistant with custom agents, document search, and access controls for multiple users.
-
codex-ppt-skill
Codex PPT Skill is a skill that turns articles, reports, and notes into image-based slide decks for people who need presentations quickly.
-
Waza
Waza is a set of eight skills that teach Claude Code, Codex, and Cursor engineering habits like planning, debugging, and reviewing code.
-
Unstract
Unstract is a self-hosted tool that pulls structured data out of PDFs and scanned documents using plain-language prompts for finance and compliance teams.
-
gemini-notebook-mcp-cli
Gemini Notebook CLI and MCP server is a command-line tool and connector that lets you manage Gemini Notebook from a terminal or AI assistant.
-
Kun
Kun is a desktop agent app that runs coding, writing, design, and research tasks on your own computer for developers and writers who want local control.
-
im-not-ai
Humanize KR is a Claude skill that rewrites AI-generated Korean text into natural Korean prose for writers and marketers.
-
Nexent
Nexent is a platform that builds AI agents from plain language descriptions, with no drag-and-drop or orchestration needed, for teams that want to self-host their agents.
-
ghidra-mcp
ghidra-mcp is an MCP server that lets AI tools analyze and document binaries in Ghidra for reverse engineers.
-
Claude Scholar
Claude Scholar is a set of skills and agents that helps computer science and AI researchers manage literature, experiments, and paper writing.
-
zotero-mcp
Zotero MCP is a connector that lets Claude, ChatGPT, and other AI assistants search, read, and annotate a Zotero research library for researchers.
-
obsidian-mind
Obsidian-mind is a self-organizing Obsidian vault that gives AI coding agents persistent memory across sessions for people who keep notes on work.
-
AIPOCH Open-Science
AIPOCH Open-Science is an agent app that helps scientists run reproducible research with AI agents, code execution, and traceable outputs on their own computer.
-
obsidian-second-brain
obsidian-second-brain is a skill plugin that gives Claude Code, Codex and other agents persistent memory stored in your Obsidian notes, for people who keep their knowledge in Obsidian.
-
OpenKnowledge
OpenKnowledge is a markdown editor that lets you edit notes and wikis with Claude, Codex, and other AI tools, stored privately on your computer.
-
Basic Memory
Basic Memory is an MCP server that gives AI chats a persistent memory stored as Markdown files for people who repeat project context every session.
-
OpenScience
OpenScience is an agent app that runs scientific research tasks like literature review, data analysis, and reproducing experiments for researchers and analysts.
-
SimpleEnglish
SimpleEnglish is a skill that makes AI agents write plain, direct technical documentation for readers who need to understand it quickly.
-
autoMate
autoMate is a self-hosted local assistant that stores your notes, files, and reminders and lets other AI apps use them for you.
-
PipesHub
PipesHub is an open-source platform that connects a company's Slack, Drive, GitHub and other business systems to AI agents, with each person's permissions respected, for teams that want self-hosted company search.
-
effective-html
Effective HTML is a set of skills that helps AI agents create self-contained HTML wireframes, prototypes, plans, and diagrams for designers and developers.
-
sepia
sepia is a skill that removes AI-sounding patterns from fiction and professional writing for writers and marketers who want text that reads as human.
-
LLM-for-Zotero
LLM-for-Zotero is an open-source research agent that lets you chat with, summarize, and organize papers in your Zotero library.
-
google_workspace_mcp
Google Workspace MCP is an MCP server that lets AI assistants like Claude control Gmail, Calendar, Docs, Sheets, Drive and other Google apps for users who run office work across Google services.
-
image-to-editable-ppt-skill
Image to Editable PPT Skill is a Codex skill that turns slide images, PDFs, and image-based PowerPoint files into editable decks for people who need to change text and layout.
-
designer-skills
Designer Skills is a collection of skills and commands that teach Claude Code and Gemini CLI to help with design work from research to delivery for designers.
-
Obsidian Local REST API
Obsidian Local REST API is an MCP server and REST API that lets AI apps and scripts read and edit your Obsidian notes for people who keep their notes in Obsidian.
-
ResumeSkills
ResumeSkills is a set of skill files that teaches Claude Code to help with resume writing, ATS checks, and job search for job seekers.
-
paper-search-mcp
paper-search-mcp is an MCP server that searches and downloads academic papers from many free sources for AI apps like Claude.
-
kordoc
kordoc is a command-line tool and MCP server that converts Korean documents (HWP, HWPX, PDF, Office files) to Markdown for people working with government and office paperwork.
-
korean-law-mcp
Korean-law-mcp is an MCP server that lets AI apps search Korean statutes, precedents, and ordinances and check legal citations for lawyers and researchers.
-
Pinvou Agent
Pinvou Agent is a desktop AI agent workspace that turns work, design, and coding requests into finished files for people who want results, not just chat replies.
-
DeepBot
DeepBot is a desktop AI assistant that handles enterprise office work such as documents, data, and monitoring for business teams.
-
webclaw
webclaw is a web extraction tool that turns websites into clean markdown or JSON for AI agents and apps, usable from a terminal, MCP, or API.
-
LocalMiniDrama
LocalMiniDrama is an open-source app that turns a written story into a short drama video on your own computer, keeping your files local.
-
Open Science Desktop
Open Science Desktop is a desktop app that runs AI research workflows from literature survey to paper, with a local, auditable record for researchers.
-
Nimbalyst
Nimbalyst is a desktop app that lets you run and visually review several coding agents at once for building software.
-
natural-japanese
natural-japanese is an Agent Skill that rewrites AI-written Japanese work documents so they read naturally for people who write them.
-
screenwriting-skills
screenwriting-skills is a set of 26 agent skills that help with writing films, TV series, and stage plays for screenwriters and writers.
-
Halo
Halo is a desktop AI workstation that runs AI team members on a schedule so work can be handed off without re-explaining it.
-
claude-fuer-deutsches-recht
Claude-fuer-deutsches-recht is a set of experimental German law skills and prompts for Claude that helps law firms and legal departments review documents and draft legal work.
-
skills
skills is a set of agent skills that guide an AI coding agent through planning, building, checking, and documenting a change for developers and vibe coders.
-
Muesli
Muesli is a native Mac app that transcribes meetings and dictation locally, with optional AI notes and agent access, for people who need meeting notes without a bot.
-
llm-wiki
llm-wiki is a plugin that turns research sources into a searchable personal knowledge base for AI coding agents like Claude Code and Codex.
-
SoloMD
SoloMD is a markdown note editor that lets AI agents like Claude Code and Cursor read and edit your notes folder for people who keep their notes as files.
-
academic-pptx-skill
Academic-pptx-skill is a skill that helps Claude build academic slide decks for conference talks, seminars, thesis defenses, and grant briefings.
-
Inno Agent
Inno Agent is a personal learning agent that remembers a learner's goals, notes, and past conversations to tailor lessons and schedule reviews for one person.
-
learn-from-materials
learn-from-materials is a skill that turns PDFs, books and papers into interactive study webpages with sourced quizzes and notes for learners.
-
JourniOne-Planning-Skills
JourniOne-Planning-Skills is a skill that turns a travel idea into a daily itinerary, a map-based travel journal, and hotel and flight comparisons for trip planners.
-
EvoOntology
EvoOntology is a plugin that builds and evolves a semantic map of your data so AI coding agents can find the right tables and meanings for data analysis tasks.
-
EasyRead
EasyRead is an app that translates English research papers into readable Chinese for people who read PDFs locally.
-
Claude-Skills-Governance-Risk-and-Compliance
Claude-Skills-Governance-Risk-and-Compliance is a set of skills that adds compliance guidance for ISO 27001, SOC 2, GDPR, HIPAA and many other frameworks to Claude for compliance teams.
-
OneBox
OneBox is an Android app with a built-in AI agent that runs its 90+ tools for you, from editing images and PDFs to bookkeeping, for everyday phone users.
-
gtrk
gtrk is a command-line video tool that lets an AI agent turn talking-head footage into editable video projects for creators.
-
academic-commercialization-agent
Academic-commercialization-agent is an agent app that turns a research paper or topic into a cited assessment of technology readiness, patents, market and risks for tech-transfer teams.
-
Agents Universe
Agents Universe is a self-hosted platform that gives teams shared AI agents with project context for getting real work done.
-
Agents Office
Agents Office is a self-hosted 3D office that routes your typed tasks to AI agents running on your own Claude login for business work.
-
OpenTakeoff
OpenTakeoff is an open-source tool that measures construction plans for bids, usable by people in a browser or by AI agents over MCP.
-
Lexora
Lexora is a desktop AI agent that works toward a goal you describe, using your files and a model you connect, for people who want help beyond coding.
-
DeepScient
DeepScient is a research workbench that guides scientists through topic selection, literature review, experiments, analysis, and writing with checkable evidence trails.
-
Agent-Me
Agent-Me is a self-hosted framework that builds an inspectable AI twin of you from your own notes and decisions, for people who want an AI that answers from their own records.
-
anki-mcp-server
An MCP server that lets AI assistants read, create, and review Anki flashcards for studying.
-
Tulpa
Tulpa is an MCP server that lets AI agents search chat history and Bilibili videos for people who work across QQ, WeChat, and Bilibili.
-
Oxynote
Oxynote is a self-hosted wiki that puts live metric charts beside the pages that explain them for teams who run services.
-
whatsapp-mcp
whatsapp-mcp is a read-only connector that lets an AI agent search and read your WhatsApp messages for you.
-
clarity
Clarity is a skill that makes drafts clearer by applying a set of writing rules when you rewrite or review prose with a coding agent.
-
Synaplan Desktop
Synaplan Desktop is a desktop app that pairs your computer with a Synaplan workspace and runs installed Agent Skills locally for people who use Synaplan.
-
gongwen-skills
gongwen-skills is a set of skill files that helps government, public institution, and state enterprise staff draft official documents without inventing facts and format them to GB/T 9704.
-
Adaptive-rag
Adaptive-rag is a self-hosted e-commerce customer service agent that answers store policy questions and handles order lookups and returns for online shops.
-
Needle
Needle is an agent app that controls an Android phone with plain English commands for people running Termux on their phone.
-
OntoPoc
OntoPoc is an agent app that drafts a business data model from spreadsheets or documents and checks it against the data, for data platform consultants.
-
dots-agent
dots-agent is an agent app that finds tasks on its own, saves reusable skills, and runs them in the background for developers and power users.
-
Best-Paper-Writing-Skill
Best-Paper-Writing-Skill is a skill that drafts research paper abstracts, introductions, methods and experiments in the style of award-winning CVPR, ICCV and NeurIPS papers for researchers writing submissions.
-
gugu
Gugu is a personal assistant app that hands tasks to AI agents that use tools and devices for people managing work and daily plans.
-
Lalabase
Lalabase is a self-hosted project management app that keeps tickets, time, docs and meetings in one place, with an AI assistant that cites its sources for software agencies and their clients.
-
Local Code
Local Code is an agent app that works through your files on Mac, iPad or iPhone, running models on the device for private work.