OpenClaw 2.0 pulls skill creation, validation, discovery, installation, and calling into one connected path, with Skill Workshop keeping the proposals, checks, and decisions on record. That is the right home for Felo skills: live web answers, X signals, decks, and transcripts that your assistant can reach from the chat app it already lives in.
bash <(curl -s https://raw.githubusercontent.com/Felo-Inc/felo-skills/main/scripts/openclaw-install.sh)
clawhub install felo-searchOpenClaw 2.0's release notes describe skills as following one connected path from creation to invocation, with Skill Workshop holding the history. Here is what that means when you install a third-party skill like Felo's.
OpenClaw meets you where you already talk — Telegram, WhatsApp, Slack, Discord, or the desktop app. These skills are chosen for what people actually ask a personal assistant to do there.
Returns dated results with sources, so a reply your assistant posts into a group can carry the link that backs it.
Search for tonight's train options from Tokyo to Osaka and send the times with sources to this chat.clawhub install felo-searchWhen the question is what people are saying right now, this reaches the posts, users, and replies instead of summarizing a news article about them.
What are developers saying about the new release today? Pull the top posts and replies.clawhub install felo-x-searchA file in the conversation beats a file in a folder. Generate the deck, then let OpenClaw's own channel tools share it under your approval rules.
Turn yesterday's research thread into a ten-slide summary deck I can send to the team.clawhub install felo-slidesImage generation is exactly the kind of job you want behind OpenClaw's approval gate: SuperAgent drafts the logo or the product image, and you decide what leaves the machine.
Generate three logo options for the side project and send them here for review.clawhub install felo-superagentAsk about a video and get the transcript, which the assistant can then summarize, quote, or turn into a deck.
Pull the transcript of this talk and list the three claims that matter for our roadmap.clawhub install felo-youtube-subtitlingOpenClaw 2.0 installs from ClawHub, then lets you decide which agents may use a skill. Pick the job you want covered: research, extraction, slides, transcripts, social monitoring, or a long-running workflow.
Add on OpenClaw 2.0
Live web research with sources you can cite
clawhub install felo-searchTurn pages into structured, reusable data
clawhub install felo-web-fetchAnswers grounded in your own documents
clawhub install felo-livedocTranscripts from talks, demos, and reviews
clawhub install felo-youtube-subtitlingTrack posts and replies on X
clawhub install felo-x-searchGenerate a deck from a single prompt
clawhub install felo-slidesTurn notes and URLs into a deck outline
clawhub install felo-content-to-slidesLong-running workflows that keep their memory
clawhub install felo-superagentCapture pages as snapshots and Markdown
clawhub install doc-screenshot-agentDraft X posts from your own material
clawhub install felo-twitter-writerGenerate a landing page from a brief
clawhub install felo-landingpageApple product research before you buy
clawhub install apple-buy-advisorPrefer the Control UI? Apply it from the Skill Workshop and enable it for the agents that should use it.
OpenClaw 2.0 keeps a record of what a skill proposed, what was checked, and who approved it, so the sequence starts small and stays reviewable.
One command per skill, and the session keeps its own transcript.
The assistant answers where the team already talks, with the link that backs it.
Recurring jobs are the point of a personal assistant, as long as a person still approves the output.
Two official routes: the installer script handles the full setup, and the ClawHub CLI fits a registry-first workflow. Both end with the skill available to your agent.
Follow the official install guide for your platform, then pair the chat app you want the assistant to live in. Skills are only useful once there is a delivery surface.
# https://docs.openclaw.ai/install
# pairing: docs.openclaw.ai/channelsThe Felo installer script copies the bundles OpenClaw expects; the ClawHub command installs from the registry when it is already configured on your machine.
# Official OpenClaw installer
bash <(curl -s https://raw.githubusercontent.com/Felo-Inc/felo-skills/main/scripts/openclaw-install.sh)
# Official ClawHub install
clawhub install felo-searchOpenClaw 2.0's Skill Workshop is the place to see what a skill proposes and what check ran before it was applied. Use it as your record.
Control UI → Skills / Skill Workshop
→ review the applied skill
→ enable it for the agents that should use itAsk a question that requires current information and check that the answer arrives with sources attached.
Search for the current status of the rail strike and reply here with two sources.These lean on 2.0's scheduling, memory, and approval model rather than treating the assistant as a chatbot.
You want a short digest in Telegram at 7:30: weather, a market signal, and anything that changed at three competitors.
Every weekday at 7:30, research these three competitors, flag only what changed, and post the brief here.A scheduled message that references previous runs instead of repeating them, with sources on every change.
The assistant should watch a topic on X, but nothing should be published under your name without approval.
Watch posts about our launch this week, summarize the recurring complaints daily, and draft replies for my approval.A daily summary plus drafts that sit behind an approval before anything reaches the channel.
A colleague asks in Slack for the comparison you researched last week, and they want slides for a meeting.
Turn the vendor comparison in my LiveDoc into a deck and post it in this thread.A deck generated from stored research and shared through the channel with an approval step.
OpenClaw 2.0 tightened approvals, credentials, and memory handling. Keep those defaults in place when you add third-party skills.
2.0 binds an approval to the exact request, command, session, and person. Skills that send messages or move files should route through that gate, not around it.
Protected credentials can reach supported destinations without entering the transcript. Store the Felo key in the CLI config and reference it by name.
Skill Workshop keeps the proposal and check history. Read it once, then decide per agent rather than enabling globally.
Memory can now be searched, inspected, imported, and removed. If a research thread should not become long-term context, keep it in a document instead.
Install one skill through ClawHub, ask a question that needs current information, and see whether the answer is worth forwarding.
Felo AI is an independent skill publisher. OpenClaw is an OpenClaw Foundation project.
Start with a skill, then compose the exact models, live data, knowledge, and output tools your OpenClaw 2.0 workflow needs — the same key every Felo skill already uses.
Use cost-efficient LLM access when your agent needs custom reasoning or generation.
Give agents current web and social signals instead of relying on stale context.
Ground work in reusable knowledge, then turn it into shareable deliverables.
Extend the workflow from research to visual and office-ready results.