Library
85,762 resources from 11 public sources. 53 checked by hand; the rest are unreviewed.
Results
-
ultrafeedback-binarized-preferences-cleaned
Datasets & benchmarksHugging Face dataset by argilla
argilla/ultrafeedback-binarized-preferences-cleaned -
stack-exchange-preferences
Datasets & benchmarksHugging Face dataset by HuggingFaceH4
HuggingFaceH4/stack-exchange-preferences -
preference-lab
Agents & frameworksPolicy-driven preference data for content moderation, with source-linked evidence, resumable runs, budget controls and TRL/ShareGPT exports.
git clone https://github.com/dengxianghua888-ops/preference-lab.git -
SHP
Datasets & benchmarksHugging Face dataset by stanfordnlp
stanfordnlp/SHP -
arena-human-preference-55k
Datasets & benchmarksHugging Face dataset by lmarena-ai
lmarena-ai/arena-human-preference-55k -
alignment-handbook
AI repositoriesRobust recipes to align language models with human and AI preferences
git clone https://github.com/huggingface/alignment-handbook.git -
orpo-dpo-mix-40k
Datasets & benchmarksHugging Face dataset by mlabonne
mlabonne/orpo-dpo-mix-40k -
Preferly
MCP serversFree AI visibility checks for any website: AI crawler access, llms.txt, schema, business info.
-
LoCo-7B
Modelstext generation model by MuyaoYuan · transformers
MuyaoYuan/LoCo-7B -
RLAIF-V-Dataset
Datasets & benchmarksHugging Face dataset by openbmb
openbmb/RLAIF-V-Dataset -
AgileRL
Agents & frameworksThe complete post-training stack for LLMs, plus RL for any problem. SFT, DPO and multi-turn agentic RL, with evolutionary hyperparameter optimization built in.
git clone https://github.com/AgileRL/AgileRL.git -
claude-reflect
SkillsA self-learning system for Claude Code that captures corrections, positive feedback, and preferences — then syncs them to CLAUDE.md and AGENTS.md.
git clone https://github.com/BayramAnnakov/claude-reflect.git -
10k_prompts_ranked
Datasets & benchmarksHugging Face dataset by data-is-better-together
data-is-better-together/10k_prompts_ranked -
bullframe.css
SkillsA semantic, lightweight CSS (PostCSS) framework for fast, accessible UIs. Class-based or classless, light or dark, agent-skills ready. Seven builds. Any stack.
git clone https://github.com/marcop135/bullframe.css.git -
claude-smart
Agents & frameworksTurns corrections into Preferences, Project-specific skills, and Shared skills for Claude Code, Codex, and OpenCode.
git clone https://github.com/ReflexioAI/claude-smart.git -
NPR Listening Service
Public APIsAudio recommendations tailored to a user's preferences
-
Local user preferences with sourced corrections, scoped recall, and Codex/Cursor host adapters. / 本地保存带来源的偏好,支持更正、限定检索与 Codex/Cursor 适配。
git clone https://github.com/KarmenLoop/user-behavior-personalization.git -
poppy
MCP serversDeveloper memory CLI and MCP server. Remember decisions, preferences, and lessons locally.
poppy-memory@0.3.6 -
Cue
MCP serversNotify people through Cue: preferences, attention budgets, quiet hours and approval apply.
cue-notify@0.6.1 -
WormaCeptor
MCP serversInspect a running Android debug app: network traffic, crashes, logs, databases and preferences
wormaceptor-mcp@3.5.0 -
memxus
MCP serversPersistent memory layer that saves and recalls your project context and preferences.
-
Agent Push Kit
MCP serversSend, search, and manage notifications, accounts, and push preferences
agent-push-kit-mcp@0.1.1 -
Phos Labs Behavioral Science
MCP serversBehavioral science tools for commerce — personas, nudges, preferences, friction audits.
-
youtube-media
MCP serversSearch YouTube, read video metadata, and fetch transcripts with language preferences
Indexed entries aren't reviewed or tested by InnerVault. Details come straight from each source. Anything that can reach your accounts, files or browser can do damage, so read the source before you install it.