Myth 1 — “You need a robotics PhD”

Reality: you need a clear state machine, a readable manifest, and a willingness to test. Most top-selling skills are written by developers with a web or mobile background who read our API docs and spent a weekend learning the simulator. Robotics credentials are useful for the hardest categories (medical-adjacent, heavy industrial); they are unnecessary for laundry, pet, hobby, and accessibility skills.

What to do instead: start with the tutorial. Build a trivial skill first. Then pick one category where you have user empathy and build for that.

Myth 2 — “You need your own compatible robot”

Reality: the GeraSkills simulator covers roughly 80% of the development loop. For the final mile (real- hardware dry-run), you either: (a) use our cloud-based physical testbench by the hour, (b) arrange a local testbench through a community makerspace, or (c) coordinate with a beta customer who owns the target hardware.

What to do instead: budget nothing for hardware until your skill passes simulator tests. Then spend £50-100 on cloud testbench time. Then, and only then, consider buying hardware.

Myth 3 — “I need to price lowest to win”

Reality: the worst thing you can do in a new marketplace is set your price by racing the cheapest listing. Buyers in 2026 are early-adopters with disposable income and they read reviews, not price tags. A well-reviewed £9.99 skill outsells a badly-reviewed £1.99 skill by orders of magnitude.

What to do instead: price against your time investment and the value delivered, not the competitive minimum. The first 100 buyers tell you whether you priced right; you can always lower.

Myth 4 — “The platform will market my skill”

Reality: we surface skills in the catalogue and promote excellent new listings, but the first 100-500 installs almost always come from the creator’s own audience (their existing social following, blog, email list, or developer community). Skills that depend entirely on platform discovery tend to stall.

What to do instead: build an audience in parallel. Post updates, share demos, write about what you’re learning. Your first customers are people who already know who you are.

Myth 5 — “Signing and review are a nightmare”

Reality: signing is one CLI command (geraskills sign), and for most skills automated review completes within minutes. Only physical-risk categories (anything involving knives, heat, navigation near children, medication) trigger human review — and even that is usually 24-48 hours. Most creators hit submit on a Thursday and are live by Friday morning.

What to do instead: skip the paranoia and submit early. If review flags something, the feedback is specific and you fix and resubmit. This is a tightening loop, not a wall.

What’s actually hard

The real hard parts, in order: (1) writing a skill that handles the messy real world, not just the happy path; (2) handling a bug report gracefully after launch; (3) staying interested enough to keep updating through months two, three, and four. Myths are distractions; these are the work.

Cross-links

Next: the tutorial, composite case study, publishing checklist. Payments across currencies: GeraCash.