The thread
Outline messagePost the outline where the discussion happened. The people who disagree will say so immediately, which is the review you wanted.
Felo Slides on OpenClaw 2.0
OpenClaw 2.0
OpenClaw 2.0A deck is more useful in the conversation than in a folder
OpenClaw can share a file in the thread it was made for, which is where a summary deck actually gets read. Felo Slides builds the outline out of the discussion you just had, so the deck is about this conversation rather than a template.
Install from the registry
clawhub install felo-slidesThis skill runs on
Read the OpenClaw 2.0 guideOpenClaw 2.0
Open-source personal assistant · release v2026.8.1
OpenClaw treats a skill as something with a lifecycle — install, review, then enable per agent — and answers wherever you already chat, so the reply has to hold up on its own.
- Registry
- ClawHub
- Review
- Skill Workshop keeps proposals and checks
- Reach
- WhatsApp, Telegram, Slack, Discord, native apps
Add it on OpenClaw 2.0From the registry to a deck your assistant can share
OpenClaw's own file handling is what delivers the deck, so the install question is only about which agents may build one and under which approval rules they may send it.
From the official registry
clawhub install felo-slidesSkill Workshop, before anything is live
- What the skill proposesThe outline step and the export step, spelled out. Read which one calls the service and which one only formats, because that decides what an approval is actually gating.
- What ran against itThe checks OpenClaw ran before applying it, kept alongside the skill so the record survives the person who installed it.
- What stays on recordApplied skills remain in the Workshop with their check history, which is the difference between an install you can audit and a folder somebody copied in.
Who gets it
Main chat agentEnabledWhere the research threads happen and where a summary is worth having while the discussion is still live.
Team channelEnabledA shared channel is the natural home for the deck, as long as your approval rule covers what leaves it.
Scheduled briefingsReview firstA recurring deck needs a fixed structure, or the audience starts reading the layout instead of the change.
The outline of the conversation you just finished having
A summary deck is only worth making while the discussion is fresh. The outline comes back as headings and one line of argument each, which is the version the people in the thread will actually read and disagree with.
Because it summarises a real thread, every line can be traced to a message somebody already sent — which is what makes the deck credible to the people in it.
Turning a long thread into something people will read
The thread already contains the work. The routine is about compressing it without losing the argument.
Runs when
A research conversation has run for an hour and the people who need the conclusion are not going to read it.
- 01
Ask for the outline in the thread
Six headings, one line each. Do it where the discussion happened so the assistant can draw on the whole thing rather than the last three messages.
- 02
Cut it in the conversation
Argue with the outline as text before anything is designed. This is the cheapest place in the whole process to change your mind.
- 03
Export last
Once the structure stops moving, have it turned into a deck. Formatting before that means formatting twice.
- 04
Share it under the rules you already set
OpenClaw's approval boundary decides what leaves the chat, so the deck inherits a policy instead of needing a new one.
Leaves behindA deck summarising the actual discussion, with the caveats preserved and the sources still traceable to messages in the thread.
Three destinations inside the same conversation
The interesting thing about a chat-first assistant is that the deck does not have to leave the place it was made.
Team channel
Shared deckSend the exported version once the argument settles, subject to the approval rule you set for that channel.
Scheduled briefing
Recurring summaryFor a recurring brief, keep the headings fixed so the audience reads the figures that changed rather than the redesign.
Two prompts that produce a deck worth sending
The first creates the structure, the second produces the file. Keeping them apart is what stops a weak claim hiding behind layout.
Summarise this thread as six slides — what we asked, what we found, what we ruled out, and what we recommend. One line each.What comes backA cuttable outline that reflects the discussion, ready to be argued with before anyone designs anything.
Turn the outline we agreed on into a deck and hold it for my approval before it goes to the team channel.What comes backA file waiting at the approval boundary, which is exactly where it should be.
What OpenClaw users ask about Felo Slides
Does the deck get sent automatically?
Only if your approval rules say so. OpenClaw's boundary governs what the agent may do with a file, and this skill produces the file rather than deciding where it goes.
Can it summarise a thread I was not part of?
It works from what the agent can see. In a channel your assistant is in, that is the conversation itself; for anything else, point it at the material explicitly.
How many slides is right?
Fewer than you think. Six headings that follow the argument beat fifteen that follow a template, and the outline stage is where that decision is cheap.
What if the thread reached no conclusion?
Then the deck should say that, with the options still open and what would settle them. A summary that invents a conclusion is worse than no summary.
Every Felo skill for OpenClaw 2.0
Five skills written for OpenClaw 2.0. They follow the same pattern on this platform, so you can combine them in one session without re-explaining your setup.
Felo Search
Felo YouTube Subtitling
Felo X Search
Felo SuperAgentOne API key. A full agent toolkit.
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.
Models
Use cost-efficient LLM access when your agent needs custom reasoning or generation.
- LLM API
- Chat completions
- Model access
Live information
Give agents current web and social signals instead of relying on stale context.
- Web Search
- X Search
- Web Fetch
Knowledge and output
Ground work in reusable knowledge, then turn it into shareable deliverables.
- LiveDoc
- PPT generation
- Mindmaps
Creative tools
Extend the workflow from research to visual and office-ready results.
- Image generation
- Page tools
- Office outputs