A list of 23 AI skills for research.
23 skills · 21–23
Anthropic's synthesize-research skill, from its product-management plugin, makes sense of a pile of user research such as interview notes, survey answers and support tickets. It groups the material into themes, ranks findings by how often they come up and how much they matter, describes user segments and suggests opportunities for the roadmap. The result is a structured research summary.
This skill uses Firecrawl to search and read the web on a complex topic, then writes a formal report with sources. The report has an executive summary, key findings, detailed analysis, contrarian views and risks, and open questions. It is meant for hard scientific, technical, policy, or market questions, not quick lookups or product picks.
This skill searches Firecrawl's paper index and writes a literature review on a topic. The index leans toward biomedical and life-science work, with PubMed, bioRxiv, and medRxiv as sources, and it also holds arXiv papers in computing, physics, and mathematics. It expands to related papers and checks claims in the paper text. The review covers key papers, themes, debates, trends, and sources.
An agent skill is a portable pack of instructions that gives an agent specialized know-how for a task. In the open Agent Skills format (from Anthropic, now widely adopted), that pack is a folder with a SKILL.md file and optional scripts or references the agent can load when needed.
SKILL.md is the required main file in a skill folder. It starts with simple metadata (name and description) plus the instructions. Agents use the metadata to decide when a skill applies, then read the full file — and any linked files — only when the task needs them.
Open the listing and follow the source link — usually a GitHub SKILL.md. Install with your agent’s skill flow, or add the repo the way that product documents. When this site shows the skill text, Copy is on the page.
A prompt is paste-in text for one chat turn (or a short thread). A skill is a reusable file pack an agent can discover and load across a task, sometimes with code attached. Prompts live in the prompt directory; skills usually live on GitHub and we link out.
Each skill uses one sitewide category — the same set as the rest of the directory (Coding agent, Marketing, Document intelligence, and so on). Hops filter to categories that already have skills.
Yes. Submit with a public GitHub SKILL.md URL, or attach a skill as part of a method on a use case. An editor reviews it before it appears here.