Research

28,063 mods in this category, of every kind an agent can take. Each one carries what it costs per session, what the scan found, and whether it is the original.

summarize

937

EvannZhongg/nanobot-desktop

Skill Claude CodeCodex

Summarize or extract text/transcripts from URLs, podcasts, and local files (great fallback for “transcribe this YouTube/video”).

not rated 53 +1 7mo ago A 32 tokens copy · 83% MIT

ethglobal

938

ethglobal-skills/repo

Plugin Claude Code

Bundles 1 skill · 56 tokens together

Search 17,000+ ETHGlobal hackathon projects, sponsor bounties, and prize winners across all events from ETHWaterloo to present.

not rated 53 5mo ago A tokens not measured

perplexity-comet-mcp

939

RapierCraft/Perplexity-Comet-MCP

MCP server Claude CodeCodexCursor +2

Bridge Claude Code with Perplexity Comet for autonomous browsing and tab management. Runs locally from the perplexity-comet-mcp npm package. Needs 2 environment variables to run.

not rated 53 4mo ago A tokens not measured original MIT

recursive-research

940

Anjos2/recursive-research

Plugin Claude Code

Bundles 1 skill · 71 tokens together

Recursive research up to PhD level across any domain (science, tech, business, arts, humanities). Source tiering, WDM + Munger inversion for autonomous decisions, disk checkpointing to survive context compaction.

not rated 53 +1 6mo ago A tokens not measured original MIT

product-deep-research

941

mingaling19/product-deep-research

Skill Codex

Run a deep multi-module research pipeline on a product or company. Output is a single complete markdown file with up to 8 modules (founders, business model, growth, use cases, ecosystem, community, competition, risks) plus optional synthesis. Use this PROACTIVELY whenever the user wants to research a product, do…

not rated 53 +1 3mo ago A 204 tokens GPL-3.0

openclaw-skill

942

Atum246/keelead

Skill Claude CodeCodex

AI-powered lead generation platform. Find leads, verify emails, research companies through natural language.

not rated 53 +5 6mo ago A 0 tokens original MIT

tap

943

vaayne/agent-kit

Skill Claude CodeCodex

Read the public web: search for sources, read one page as Markdown, and run site scripts for structured records (X/Twitter via FxEmbed, Exa search, GitHub, Hacker News, Reddit, Wikipedia, Bilibili bundled; a catalog of more installable with one command). Use for any task that needs something from a public website: a…

not rated 53 yesterday A 116 tokens original MIT

timecode-agent

944

mupozg823/timecode-agent

Skill Codex

A workflow for understanding and indexing existing videos using timestamps, transcripts, visual checks, and reusable notes.

not rated 52 3mo ago A SkillSpector: pass 113 tokens original MIT

xvhaoran778-cyber/Serenity.SKILL

Skill Codex

Cross-market supply-chain chokepoint investing research workflow inspired by Serenity's public posts and trade disclosures. Use when analyzing A-share, US, Hong Kong, Taiwan, Japan, Europe, or other listed stocks, sectors, or trade ideas through hidden bottlenecks, AI/semiconductor/energy infrastructure supply chains…

not rated 52 5mo ago A 148 tokens original MIT

litllm

947

LitLLM/LitLLM

Skill Claude Code

Generate a related-work section, find prior literature, or rank candidate citations for an academic paper. Uses LLM-driven debate ranking over Semantic Scholar / arXiv / OpenAlex search, with optional citation-graph BFS expansion (deep research). Use when the user provides a paper (PDF or markdown) and asks for…

not rated 52 6mo ago C 80 tokens original Apache-2.0

research-proof-plugin

948

tonyblu331/research-proof

Plugin Claude Code

Bundles 1 skill · 101 tokens together

Falsifiable research planning for AI agents: pressure-test claims, freeze evaluators, use AI-lab and medical evidence patterns, build proof ladders, and maintain proof ledgers.

not rated 52 5mo ago A tokens not measured original MIT

deepsrt-mcp

949

DeepSRT/deepsrt-mcp

MCP server Claude CodeCodexCursor +2

DeepSRT MCP Server - YouTube video summarization and transcript extraction. Runs locally from the @deepsrt/deepsrt-mcp npm package.

not rated 52 1y ago A tokens not measured

scopus-mcp

950

qwe4559999/scopus-mcp

MCP server Claude CodeCodexCursor +2

A Model Context Protocol server for the Elsevier Scopus API. Runs locally from the scopus-mcp Python package. Needs 1 environment variable to run.

not rated 52 +3 5mo ago A tokens not measured original MIT

gvkhosla/compound-engineering-pi

Skill Claude CodeCodex

Fetches and analyzes GitHub issues to surface recurring themes, pain patterns, and severity trends. Use when understanding a project's issue landscape, analyzing bug patterns for ideation, or summarizing what users are reporting.

not rated 52 1mo ago A 49 tokens original MIT

hn

952

heartleo/hn-cli

Plugin Claude Code

Bundles 1 skill · 149 tokens together

Hacker News for Claude. Scan the front page for today's themes, hot discussions, and industry mix.

not rated 51 2mo ago A tokens not measured original MIT

khanote.start-my-day

953

almost-khan/khanote

Skill Claude CodeCodex

You are running the daily research briefing or ad-hoc research workflow for the user.

not rated 51 7mo ago A 0 tokens

LiXin97/agora-lab

Skill Claude CodeCodex

Design experiments with variables, baselines, ablations, resources, and success criteria.

not rated 51 6mo ago A 22 tokens original Apache-2.0

amirkiarafiei/open-scholar-peer

Skill Claude CodeCodex

Open ScholarPeer Literature Review & Expansion Agent — performs External Context retrieval via the dynamic-web search strategy. Activate this persona when the user invokes /2-osp-literature. Runs up to three distinct rounds — the user chooses how many (sub-domain anchor, method anchor, temporal expansion) to construct…

not rated 51 changed yesterday A SkillSpector: pass 72 tokens original MIT

research-craft

957

nik1t7n/research-craft-skill

Skill Codex

Use when planning, reviewing, or improving research work: choosing research problems, designing ML/AI experiments, reading papers, building experiment loops, analyzing outputs, writing research notes, or turning vague curiosity into a disciplined research plan. Emphasizes avoiding absorbed problems, upgrading…

not rated 50 4mo ago A 76 tokens original MIT

xueqiu-summary

959

chenmuwen0930-rgb/openclaw-skill-snowtrace

Skill Claude Code

A Chinese-language skill that gathers recent posts from followed Snowball influencers and market data for a user's watchlist. Snowball is a Chinese investment social network, and the skill combines the results into an investment reference report.

not rated 50 8mo ago A 40 tokens original MIT

llm-wiki

960

Oshayr/LLM-Wiki

Plugin Claude Code

Bundles 5 skills, 10 agents · 634 tokens together

LLM-powered personal wiki — autonomous knowledge base with research-on-miss, ingestion, search, a browsable web UI, and universal data gravity. Saves and retrieves knowledge automatically whenever relevant.

not rated 50 6mo ago A tokens not measured original MIT

At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: