Show HN: Codified decades of domain expertise into open source agent skills
2 points| urav | 4 days ago |github.com
Evos started out as doing some AI consulting, while deploying these systems, I kept hitting the same problem: AI agents are terrible at real operational work in core industries. Agents could code the entire system, but asking it to handle a freight exception, or claims deadline issue and it fell apart - it just didn't have the domain knowledge.
So I started translating and codifying what operations experts actually know - the decision frameworks, edge cases, escalation protocols - into agent skills for the Evos platform.
Last week, I noticed there are over 3000 skills on ClawHub, and nearly all of them are dev tool wrappers and API integrations. As far as I could tell, there aren't many skills that teach agents genuine domain expertise for traditional industries - and even fewer coming from verified expertise.
Today, I open sourced the Evos agent skills.
We've published 8 skills for use cases across logistics, manufacturing, retail and energy. Each of them follow the Agent Skills open standard, and work across all the major platforms.
Beyond the skills, I built in an eval suite - 20 to 25 scenarios from real operations, scored against weighted rubrics and also benchmarked agents with the skills vs agents without.
Repo: https://github.com/ai-evos/agent-skills
Would love your feedback from anyone whose been in these industries, or thoughts as a whole.
I'll be parked in the comments - thanks!
urav|1 day ago
Startups as a whole have been focused on building for agent use, assuming that a future of autonomous work is just going to emerge - yet aside from tech and engineering - claude code or openclaw isn't going to be running logistics ops, or admin for healthcare teams...
that's 70% of the worlds workforce, in legacy industries that are still being given a copilot, instead of a solution that actually reduces the workload, and frees up time.
i feel like that barrier is domain expertise. i mean i'm barely 23 years old, what do I, as a dev, know about ops challenges in manufacturing?
but with evos, by translating and codifying that expertise - it not only allows us to build domain specific applications, with a high success rate - and allow agents to carry niche expertise of a specific role, in a specific company - and let's an autonomous system start as if it's been an employee in the company for 30 years.
Maybe sometime in the future, these will be capabilities of domain expertise that we can plug into a neuralink... how sick would that be?
a rant, but the chain of thought is there. would love to discuss other's opinions