SkillWorks

Skills

Subagents

Plugins

Marketplaces

SkillWorks

Skills

Subagents

Plugins

Marketplaces

Submit a repo

SkillWorks

Skills

Subagents

Plugins

Marketplaces

Submit a repo

Skills/AI & agents/K-Dense-AI/scientific-agent-skills

arbor

K-Dense-AI/scientific-agent-skills
SkillLoads cleanly

Autonomously improve a real artifact (code, training recipe, agent harness, data pipeline, prompt) against an objective and an evaluator, using Hypothesis…

Scores
89
out of 100 · grade A

2026-08-22

Works

40% of the score

92/100

  • Names 1 tool Claude Code does not ship: Read Write Edit Bash Agent.

Maintained

25% of the score

100/100

  • no commits in the last 12 weeks

Adopted

20% of the score

62/100

  • 34,091 stars on the source repo.

Documented

15% of the score

100/100

  • 1,871 words with worked examples.
  • Ships 5 bundled files.
Loads cleanly. Every structural check Claude Code needs to register this passed. Last commit 2 days ago.

Install

npx skills add K-Dense-AI/scientific-agent-skills/arbor

What the check found

Every finding on this listing, and what each one means for whether it loads.
FindingWhat it means
Unknown toolsIt allow-lists tool names Claude Code does not ship, which usually means a typo.

What it says it does

Autonomously improve a real artifact (code, training recipe, agent harness, data pipeline, prompt) against an objective and an evaluator, using Hypothesis Tree Refinement (HTR) from the Arbor paper. Use this whenever someone wants to iteratively optimize something over many experiments without overfitting — e.g. "get my model's eval score up", "improve this agent/harness", "tune this pipeline", "beat the baseline on this benchmark", "run a search over approaches and keep the best", "do an MLE-bench / Kaggle-style optimization", or any long-horizon "make this artifact better and don't just memorize the dev set" task. Trigger it even when the user doesn't say "Arbor" or "hypothesis tree" but describes repeated experiment-and-evaluate loops, branching exploration of competing ideas, or worries about a dev/test gap. Runs Claude itself as the coordinator with subagent executors in isolated git worktrees; for the standalone `arbor` CLI tool see references/arbor-upstream.md.

Every number above came out of reading the file and its repository. Nothing is a judgement call, nothing is a model’s opinion, and nothing can be paid for. The full method is published. Source last committed 2026-08-19.

The record

Type
Skill
Category
AI & agents
Repository
K-Dense-AI/scientific-agent-skills
Last commit
2 days ago · 2026-08-19
Repository stars
34k
Installs
not in the skills.sh registry
Licence
MIT
Tools it asks for
Read Write Edit Bash Agent
Instruction length
1,871 words
Bundled files
5 · includes scripts
View the source on GitHub

Also in K-Dense-AI/scientific-agent-skills

Other artifacts published from the same repository.
ArtifactScoreWhat the check foundTypeReachLast commit
arboretoK-Dense-AI/scientific-agent-skills
92
cleanSkill1,094 installs2 days ago
bidsK-Dense-AI/scientific-agent-skills
92
cleanSkill34k stars2 days ago
clinical-decision-supportK-Dense-AI/scientific-agent-skills
92
cleanSkill1,111 installs2 days ago
dnanexus-integrationK-Dense-AI/scientific-agent-skills
92
cleanSkill1,078 installs2 days ago
get-available-resourcesK-Dense-AI/scientific-agent-skills
92
cleanSkill1,171 installs2 days ago
imaging-data-commonsK-Dense-AI/scientific-agent-skills
92
cleanSkill1,101 installs2 days ago

Other ai & agents skills

Browse all
Same category, same type, ranked by score.
ArtifactScoreWhat the check foundCategoryReachLast commit
mcp-builderanthropics/skills
97
No licenseAI & agents104,949 installstoday
agent-developmentanthropics/claude-plugins-official
96
cleanAI & agents5,901 installstoday
plugin-settingsanthropics/claude-plugins-official
96
cleanAI & agents5,465 installstoday
skill-creatoranthropics/claude-plugins-official
96
cleanAI & agents5,836 installstoday
microsoft-foundrymicrosoft/azure-skills
96
cleanAI & agents544,819 installstoday
claude-apianthropics/skills
95
No licenseAI & agents59,236 installstoday

Put this measurement in your README

A badge carrying how many listings this index holds from the repository and how many pass every static structural check. It reads from this index every time somebody loads your page, so it changes when the measurement changes and there is nothing to keep up to date. Free, no account, and the value is not something you or we can set by hand.

SkillWorks badge

[![SkillWorks](https://toolproof.kynth.studio/badge/skillworks/K-Dense-AI/scientific-agent-skills.svg)](https://skillworks.kynth.studio/?q=K-Dense-AI%2Fscientific-agent-skills)

Would rather not hotlink us? Every badge is also served in shields.io’s endpoint schema, so shields renders the image and your readers never talk to our domain:

Published by Toolproof, the masthead over this index and eight others. The method behind the number is at toolproof.kynth.studio/methodology, and the whole thing is readable as JSON with no key at /api.

Also from Kynth Studios

Built for the same person as SkillWorks

Toolproof

Independent measurement of AI agent tooling

toolproof.kynth.studio

BlockDex

Search public shadcn registries, item by item

blockdex.kynth.studio

KitGrade

SaaS starter kits, scored on what can be checked

kitgrade.kynth.studio

The studio list

One product, taken apart, once a month

Kynth Studios pulls one shipped product open every month — what it does, what it cost to build, what the pipeline behind it looks like, and what the numbers did. One email a month, nothing in between.

Double opt-in — we send one confirmation link and nothing else until you click it.

SkillWorks

Built by

Kynth Studios

the studio behind Toolproof, BlockDex and KitGrade

part of Toolproof, the measurement layer for AI agent tooling

Browse

Skills
Subagents
Plugins
Marketplaces
CLAUDE.md examples

Rankings

Best subagents
Best plugins
Most installed
Will not load

The index

Search
Categories
How we score

Reference

Submit a repository
Sponsor a section
Listings JSON
Privacy

© 2026 SkillWorks. A Kynth Studios product. Changelog

SkillWorks

The studio list

One product, taken apart, once a month

Kynth Studios pulls one shipped product open every month — what it does, what it cost to build, what the pipeline behind it looks like, and what the numbers did. One email a month, nothing in between.

Double opt-in — we send one confirmation link and nothing else until you click it.

SkillWorks

Built by

Kynth Studios

the studio behind Toolproof, BlockDex and KitGrade

part of Toolproof, the measurement layer for AI agent tooling

Browse

Skills
Subagents
Plugins
Marketplaces
CLAUDE.md examples

Rankings

Best subagents
Best plugins
Most installed
Will not load

The index

Search
Categories
How we score

Reference

Submit a repository
Sponsor a section
Listings JSON
Privacy

© 2026 SkillWorks. A Kynth Studios product. Changelog

SkillWorks

The studio list

One product, taken apart, once a month

Kynth Studios pulls one shipped product open every month — what it does, what it cost to build, what the pipeline behind it looks like, and what the numbers did. One email a month, nothing in between.

Double opt-in — we send one confirmation link and nothing else until you click it.

SkillWorks

Built by

Kynth Studios

the studio behind Toolproof, BlockDex and KitGrade

part of Toolproof, the measurement layer for AI agent tooling

Browse

Skills
Subagents
Plugins
Marketplaces
CLAUDE.md examples

Rankings

Best subagents
Best plugins
Most installed
Will not load

The index

Search
Categories
How we score

Reference

Submit a repository
Sponsor a section
Listings JSON
Privacy

© 2026 SkillWorks. A Kynth Studios product. Changelog

SkillWorks