By pixeltable
Build multimodal AI applications with Pixeltable -- tables, computed columns, embedding search, UDFs, tool-calling agents, and 25+ AI provider integrations.
Scaffold a new Pixeltable project from an official template or structural pattern.
Build a document RAG pipeline in Pixeltable (chunk, embed, retrieve, answer).
Add an AI provider call as a Pixeltable computed column (correct import + output shape).
Diagnose a failing or stale Pixeltable pipeline (errored columns, wrong logic, retrieval issues).
Builds end-to-end RAG and semantic-search pipelines in Pixeltable — chunking, embedding indexes, retrieval, and grounded answer generation. Use when the user wants to make documents, images, audio, or video searchable and answerable without a separate vector DB or framework.
Designs Pixeltable schemas — tables, views/iterators, computed-column chains, embedding indexes, and UDFs — for multimodal and ML data pipelines. Use when the user needs to model a data/AI workflow or decide between a view, a computed column, and a UDF.
Diagnoses and fixes failing or stale Pixeltable pipelines — errored computed columns, no-op recomputes, retrieval problems, rate limits, and deprecated-API misuse. Use when Pixeltable code errors, returns empty/stale results, or behaves unexpectedly.
Modifies files
Hook triggers on file write and edit operations
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Agent Skill that teaches AI coding assistants to write correct Pixeltable code — declarative tables that replace LangChain + pandas + vector databases with one system. Covers 25+ AI providers, multimodal pipelines, tool-calling agents, RAG, and production patterns.
Which path? Use npx plugins add for the full plugin (skill + agents + slash commands) on Claude Code and Cursor. Use npx skills add to install just the skill content across 40+ agents (Copilot, Windsurf, Gemini, etc.) or in CI. Both work; they're complementary.
npx plugins add pixeltable/pixeltable-skill
npx skills add pixeltable/pixeltable-skill
/plugin marketplace add pixeltable/pixeltable-skill
/plugin install pixeltable@pixeltable-skill
codex plugin marketplace add pixeltable/pixeltable-skill --ref main
codex plugin add pixeltable@pixeltable-skill
skills/pixeltable-skill/ # The skill (content core)
├── SKILL.md # Core: negative prompts, task router, API, agents, pitfalls
└── references/ # Loaded on demand by Claude Code / Cursor
├── core-api.md # Tables, querying, views, UDFs, config, data sharing
├── providers.md # 25+ AI providers with quick-reference table
├── workflows.md # RAG, video, image, audio, FastAPI, export
├── video-rag-agents.md # Video + transcript search + agent
├── agents-memory-mcp.md # Agent with memory, MCP, multi-provider
├── ml-data-pipeline.md # Ingest, enrich, version, PyTorch export
├── agentic-patterns.md # 6 patterns + 2 reasoning strategies
└── anti-patterns.md # 15 training-distribution biases with wrong/right code
commands/ # Slash commands: /pixeltable:scaffold, rag, add-provider, debug
agents/ # Specialists: rag-builder, pipeline-architect, debugger
hooks/ # Optional pure-Python hooks (Claude Code): orientation + anti-pattern validation
The plugin install (npx plugins add) bundles the skill, commands, agents, and hooks. The skill install (npx skills add) delivers just the skill content. Hooks run on Claude Code; Cursor honors session-start context only.
See CONTRIBUTING.md. The skill content lives in skills/pixeltable-skill/; run python3 scripts/validate_plugin.py after structural changes.
uvx pixeltable-new myapp (pixeltable-new)uvx pixeltable-new --template <name> my-app — knowledge-base, chat-agent, audio-transcription, video-search, media-indexing, image-dataset, full-stack-showcaseApache 2.0
npx claudepluginhub pixeltable/pixeltable-skill --plugin pixeltableMulti-model consensus engine integrating OpenAI Codex CLI, Gemini CLI, and Claude CLI for collaborative code review and problem-solving.
Ultra-compressed communication mode. Cuts ~75% of tokens while keeping full technical accuracy by speaking like a caveman.
Comprehensive UI/UX design plugin for mobile (iOS, Android, React Native) and web applications with design systems, accessibility, and modern patterns
Curate auto-memory, promote learnings to CLAUDE.md and rules, extract proven patterns into reusable skills.
Memory compression system for Claude Code - persist context across sessions
Standalone image generation plugin using Nano Banana MCP server. Generates and edits images, icons, diagrams, patterns, and visual assets via Gemini image models. No Gemini CLI dependency required.