A package manager for agent skills with built-in evals
Details
- External ID
- 46900933
- Source
- HN
- Company
- —
- Product
- A package manager for agent skills with built-in evals
- Website domain
- tessl.io
- Launched
- Feb. 5, 2026
- Cohort
- —
- Upvotes
- 7
- Upvotes percentile
- 0.38207547169811323
- Tags
- —
- Fetched at
- Sept. 7, 2026, 9:26 p.m.
- Updated at
- Sept. 7, 2026, 9:26 p.m.
Description
I'm Guy, the founder behind Snyk — now building Tessl, a package manager for agent skills.We’ve recently witnessed that most teams still treat skills as static artifacts: markdown files, created or copied from repo to repo.This approach offers a strong initial boost, but quickly creates debt:- Skills are duplicated, and updates never roll out. - Poor quality skills go unseen, misguiding agents instead of helping. - Skill knowledge grows stale, and don’t keep up with the systems and practices they describe.Without a way to evaluate skills, teams have no clear way to understand how good a skill actually is, or if it degraded over time.Our belief is that evaluations are the foundation for having quality skills.With that in mind, I’m glad to announce that Tessl Registry contains review evals for over 2,000 skills, and you can request an evaluation for any public skill.Super excited to be launching this — keen to get your feedback, and looking forward to the many more enhancements in the queue!
Enrichment
- Theme
- AI agent frameworks and developer tools
- Vertical
- Horizontal
- Function
- Dev tools
- Audience
- Developer
- AI stance
- AI-native
- Project type
- Commercial product
- Normalized one-liner
- package manager for ai agent capabilities
- Manually corrected
- False
Could you build this?
No Building an enterprise-grade package manager and evaluation infrastructure for agent skills requires complex automated sandbox environments, rigorous statistical benchmarking against nondeterministic models, and package registry architecture.
What it would actually take: A real platform requires a distributed registry with semantic versioning and dependency resolution, combined with isolated execution sandboxes (Firecracker microVMs or secure Docker containers) to run agent evals safely. The difficult core is designing an evaluation harness that executes diverse LLMs against agent skills, handles nondeterministic tool-calling evaluations, computes reliable benchmarks across model updates, and manages multi-tenant registry security. It demands deep domain expertise in compiler/package manager architecture, LLM evals, and secure multi-tenant infrastructure.
Discussion
2 comments analyzed.
Concerns raised: Model changes cause prompt/skill regressions over time, Difficulty tracking skill quality and drift across versions
Feature requests: CI integration to pin skill versions and fail builds on eval score drops, Skill quality visibility over time with regression detection
Competitors
Other products that read as similar to this one — 197 launches clear the similarity bar, closest 8 shown.
Attention rank: #124 of 198 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 99 days after the earliest competitor.
- HyperSkills · github · 2026-09-10 · 22 upvotes · similarity 0.48
- Agent-evals · hn · 2026-05-04 · 9 upvotes · similarity 0.47
- A registry for curated, high quality Claude skills and skillsets · hn · 2026-01-22 · 9 upvotes · similarity 0.46
- TLA+ Workbench skill for coding agents (compat. with Vercel skills CLI) · hn · 2026-02-22 · 41 upvotes · similarity 0.44
- Distributing AI agent skills via NPM · hn · 2026-01-09 · 6 upvotes · similarity 0.43
- Webhook Skills · hn · 2026-02-04 · 9 upvotes · similarity 0.42
- Ratel, give agents unlimited tools and skills without context bloat · hn · 2026-07-16 · 23 upvotes · similarity 0.42
- Skillscript · hn · 2026-07-12 · 18 upvotes · similarity 0.42
Other launches for this product
- No other launches for this product.
Same idea, different domain
Nobody's really built a dev tools tool for Sales yet.