LLMRepos Weekly
Five repos worth watching, every Tuesday.
The weekly index
Weekly editions, at a glance
Five repos per edition, with each repo's star gain and % change over the edition's seven-day window.
Does that skill actually help your agent?
NVIDIA tests whether skills improve results, Univer gives agents editable spreadsheets, and Jane Street shares a tutor that makes you do the thinking.
Repos covered · stars gained in 7 days, % change
- NVIDIA/SkillEvaluator+2,749+409%
- dream-num/univer+6,510+28%
- alibaba/open-code-review+4,247+8.5%
- janestreet/agent_skills+5,698+134%
- HandsOnLLM/Hands-On-Large-Language-Models+86+0.3%
Find the file you only half remember
Kensaku searches inside your files locally, NVIDIA's Auto Ontology tells agents what your data means, and DeepSeek publishes an agent that rations attention.
Repos covered · stars gained in 7 days, % change
- yjid/kensaku+955+184%
- NVIDIA/auto-ontology+10,152+1,414%
- D4Vinci/Scrapling+7,926+8.6%
- deepseek-ai/dsh-superagent-prototype+9,772+82%
- Coding-Solo/godot-mcp+105+1.6%
Let your agent read the desktop
Agent Desktop targets named controls, Dossier tailors job applications, and DeepTutor turns your documents into study material.
Repos covered · stars gained in 7 days, % change
- AienTech/dossier+3,583+675%
- vercel-labs/agent-desktop+9,199+504%
- HKUDS/DeepTutor+10,059+23%
- google/elephant-goldfish-model+8,937+473%
- SakanaAI/AI-Scientist+121+0.8%
Make your agent challenge its own findings
Cloudflare separates security hunters from verifiers, ThinkRail gives agents separate workspaces, and Lemo-Opuscar makes films in code.
Repos covered · stars gained in 7 days, % change
- JetBrains/thinkrail+4,424+452%
- experientiallabs/experiential+3,520+35%
- cloudflare/security-audit-skill+14,753+40%
- lemomo-ai/lemo-opuscar+4,180+83%
- GeeeekExplorer/nano-vllm+150+0.9%
Give your agent something you can inspect
SolidCraft makes editable CAD parts, Anatomy builds interactive explanations, and DigUp searches inside your files.
Repos covered · stars gained in 7 days, % change
- tangency-labs/solidcraft+5,590+636%
- wheresryan22/anatomy+11,457+226%
- K-Dense-AI/scientific-agent-skills+4,294+8.8%
- ARahim3/DigUp+3,265+199%
- mlabonne/llm-course+818+1.0%
Ask your agent how that app works
REA looks inside apps, Concat gives agents a video timeline, and XERJ helps them find the right code.
Repos covered · stars gained in 7 days, % change
- morluto/rea+239,845+442%
- jub0t/Concat+3,261+72%
- xerj-org/xerj+3,636+105%
- yetone/magpie+6,606+85%
- shiyu-coder/Kronos+448+1.1%
Give your agent a lighter browser
Moli skips rendering until you need it, plus persistent agent teams, a self-hosted assistant and local voice workflows.
Repos covered · stars gained in 7 days, % change
- mvschwarz/openrig+3,859+330%
- spinabot/brigade+4,145+58%
- debpalash/VoiceStudio+12,460+30%
- lexmount/moli+5,435+217%
- dair-ai/Prompt-Engineering-Guide+148+0.2%
Start with your agent's failures
An eval workflow worth trying first, plus agent memory, team coordination, API tooling and a design guide.
Repos covered · stars gained in 7 days, % change
- ai-evals-course/evals-skills+731+115%
- vectorize-io/hindsight+16,799+68%
- paperclipai/paperclip+12,408+15%
- cortex-docs/cortex+3,072+1,920%
- humanlayer/12-factor-agents+102+0.4%
Treg doubled its stars in a week
A tool gateway doubles, a video workflow climbs, and DeepSeek's new agent harness draws attention.
Repos covered · stars gained in 7 days, % change
- superdesigndev/treg+1,741+104%
- hypit-ai/hypit+5,777+56%
- stablyai/orca+6,366+8.8%
- deepseek-ai/deepseek-harness+6,875+3.0%
- karpathy/autoresearch+476+0.5%
Prefer the weekly email?
Five repos and one short read every Tuesday.
Free weekly digest. Double opt-in, unsubscribe anytime.