Most OpenClaw agents are brilliant at reasoning, writing, and working with local files — but ask one what happened in the news an hour ago and you’ll hit a wall. That’s the gap Felo AI set out to close. Their recently launched suite of eight interconnected ClawHub skills turns any OpenClaw agent into a real-time research assistant with proper source citations, multilingual support, and a few unexpected extras like slide generation and social media monitoring.
What Is Felo Search?
Felo AI is a search engine built for the AI-native era. Unlike traditional search tools that return ranked lists of blue links, Felo synthesizes information from multiple sources and returns a structured answer with inline citations. Think of it as giving your agent the ability to do what a skilled researcher does: read multiple sources, cross-reference them, and produce a coherent summary with references.
The OpenClaw integration packages this capability — and several others — into a set of skills you can install from ClawHub with a single command:
npx clawhub@latest install felo-ai/felo-searchA free API key from Felo unlocks all eight skills in the suite. No paid tier is required to get started.
The Eight Skills, Explained
Felo’s ClawHub offering is not a single monolithic skill but a modular suite. Here’s what each piece does and when your agent will reach for it.
1. Felo Search
The core skill. Your agent sends a natural-language query and gets back an AI-synthesized answer with numbered source citations. Results are drawn from the live web, so the information is current — not limited to a training cutoff. Multilingual queries are supported natively: ask in Japanese, get results that pull from both Japanese and English sources.
2. Felo SuperAgent
A more powerful research mode that performs multi-step reasoning. SuperAgent breaks complex questions into sub-queries, searches each independently, and merges the results into a comprehensive report. Ideal for competitive analysis, literature reviews, or any task where a single search query isn’t enough.
3. Felo Web Fetch
Retrieves and parses the full content of a specific URL. When your agent already knows where to look — a documentation page, a press release, a regulatory filing — this skill pulls the text cleanly without boilerplate navigation and ad content.
4. Felo X (Twitter) Search
Searches posts on X (formerly Twitter) and returns results with author handles, timestamps, and engagement metrics. This is particularly useful for tracking real-time reactions to product launches, monitoring brand mentions, or gathering sentiment data that doesn’t appear on traditional search engines.
5. Felo YouTube Subtitles
Extracts subtitle text from YouTube videos. Give your agent a video URL and it returns the transcript, which can then be summarized, searched, or used as source material. Handy for podcast summaries, conference talk notes, or extracting quotes from interviews.
6. Felo LiveDoc
Creates a living document from a search query that updates as new information becomes available. Think of it as a persistent research brief that stays current — useful for ongoing projects where the landscape changes frequently, like tracking a regulatory proposal or a developing news story.
7. Felo Slides (Content-to-Slides)
Converts research output into presentation slides. Your agent can search for information, synthesize it, and then generate a slide deck — all within a single workflow. The output is structured with headings, bullet points, and speaker notes.
8. Felo Content-to-Slides (Advanced)
An extended version of the slides skill that accepts arbitrary text or documents as input, not just Felo search results. Feed it a report, a set of meeting notes, or a product spec, and it generates a polished slide outline.
Who Is This For?
The Felo suite is particularly well-suited for three types of OpenClaw users. First, researchers and analysts who need agents that can pull current information with verifiable sources rather than hallucinating facts from stale training data. Second, marketing and communications teams who want to monitor social media, track competitors, and turn findings into presentations without leaving their agent workflow. Third, developers building agent pipelines who need a reliable, citation-rich search primitive to chain with other skills.
The multilingual capability also makes this a standout for international teams. Agents can search across language barriers — querying in one language and retrieving relevant results from sources in others — which is something most search-oriented skills on ClawHub don’t handle well yet.
Getting Started
Installation is straightforward:
npx clawhub@latest install felo-ai/felo-searchAfter installing, you’ll need to register for a free API key at felo.ai and add it to your OpenClaw configuration. The free tier is generous enough for personal use and prototyping. For production workloads with high query volumes, Felo offers paid plans with higher rate limits.
You can find the full documentation and source on the Felo OpenClaw skills launch page and on the Sundial curated skills list, where Felo Search is categorized under Search & Research.
The Bigger Picture
With ClawHub now the default skill registry as of OpenClaw v2026.3.22, the barrier to adding capabilities like real-time search has dropped to a single terminal command. Felo’s suite is a good example of what that unlocks: a third-party company packaging a sophisticated API into modular, composable skills that any OpenClaw agent can use out of the box. As the ecosystem matures, expect more specialized search and data providers to follow the same playbook.
Leave a Reply