Several people asked the same onboarding question and got no answer from the official account.
Unanswered questionFelo X Search on WorkBuddy
WorkBuddy
WorkBuddyWhat customers say where the brand is not listening
Support tickets arrive from people who bother to write in. Posts arrive from people who just talk. Felo X Search turns that chatter into a short list of things worth acting on, with the handle attached.
No terminal needed
Skills → Add skill → Upload skill → felo-x-searchWorkBuddy runs a bench of specialists on your desktop, in WeChat and WeCom. Skills arrive as packages with a permission list, so research becomes a capability the whole bench can call.
- Runs on
- Desktop, WeChat, WeCom, QQ, mini program
- Skills arrive by
- Package upload, with permissions listed
- Fits
- Founders, research and operations teams
A short list, grouped by what it implies
Raw social results are unusable in a report. The skill groups what it finds by what it suggests you do — a complaint pattern, a competitor move, a question nobody answers.
A competitor's new tier is being compared against ours on price, usually unfavourably.
Competitor moveThe same setup step is described as confusing by people who otherwise praise the product.
Friction patternA feature request we already shipped keeps coming up, which suggests the announcement was missed.
Messaging gapAsk for counts alongside quotes: four posts from four people is a different finding from four posts from one thread.
Add it on WorkBuddyImport it, then decide who may run it
Because this skill reads public conversation about your brand, the permission conversation matters more than the install. WorkBuddy shows you the list before it turns on.
- 01
Take the skill folder
Get the felo-x-search folder from the repository. It contains the manifest and the scripts the skill runs.
- 02
Import under Skills
Skills → Add skill → Upload skill. Read the permission list carefully — this skill has no business asking for account access.
- 03
Enable it for the people who report on this
Social findings are easy to misread out of context. Keep it enabled for the team that knows the history, and share the output as a document rather than a live tool.
Then ask for a count, not just quotes
Search X for what people are saying about our onboarding.
Group the findings and tell me how many people each one comes from.Two reports worth running monthly
Both are short, and both are better as a recurring habit than a one-off panic search.
What are people complaining about this month that we have not answered publicly? List the complaints and whether anyone replied.What comes backA short list of gaps where the brand is absent, which is usually more actionable than a sentiment score.
Compare how customers describe us against how they describe the two named competitors, using their own words.What comes backThe language customers actually use, which is more useful for messaging than any summary you write yourself.
Turn chatter into something with a number attached
Social data tells you what to investigate. These two help you finish the sentence.
Sense-checking a social finding
Social data fails by being loud rather than wrong. These checks keep a finding proportional.
Count the people, not the posts
One angry thread can look like a movement. Ask how many distinct accounts are behind a finding before it goes in a report.
Check the date range
A result from eight months ago reads like this week unless you ask. Confirm the window matches the period you are reporting on.
Find one corroborating source
If the finding matters, look for a support ticket, a review, or a search result that says the same thing somewhere other than X.
What WorkBuddy users ask about social data
Is this safe to run inside a company workspace?
It reads public data through the Felo API and needs no social account, so nothing about your organisation's credentials is exposed. The findings are public information, which is worth remembering before they are forwarded.
Can it monitor a specific competitor continuously?
It can run whenever you ask. For a standing watch, pair it with Felo SuperAgent so the checking happens on a schedule and only the changes reach you.
Why is the output grouped rather than listed?
Because a hundred posts is not a finding. Grouping by implication — unanswered question, competitor move, friction pattern — is what makes the result short enough to act on.
Every Felo skill for WorkBuddy
Five skills written for WorkBuddy. They follow the same pattern on this platform, so you can combine them in one session without re-explaining your setup.
Felo Search
Felo Slides
Felo YouTube Subtitling
Felo SuperAgentOne API key. A full agent toolkit.
Start with a skill, then compose the exact models, live data, knowledge, and output tools your WorkBuddy 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