Reusable AI skills and capabilities for agent workflows
Yufeng He <[email protected]>
Generate AI assistant rules (CLAUDE.md, .cursorrules, copilot-instructions) from codebase analysis
Nidesh Pandey
Lightweight terminal tool to check remaining credits / quota across LLM providers. Zero dependencies.
Anish Shrestha
Relevance-driven CPU KV-cache offload policy for stock vLLM
Agentix-E
OpenAI-compatible LLM provider for @agentix-e/nl2spel
kevinhao
AI SDK v6 provider for Pi Coding Agent (supports 15+ LLM providers)
specllm contributors
Makes any LLM behave like a regular REST API. Define the contract, specllm enforces it.
aimedia.az
AIMEDIA.AZ AI — çox-provayderli, failover + retry-li AI Chat Model node (NVIDIA, DeepSeek və istənilən OpenAI-uyğun endpoint).
...morecircuit-llm
Circuit SDK core — http client, injectable config, ed25519 identity, shared types.
mehrdadashtari
Cloudflare Workers AI nodes for n8n, starting with a Cloudflare Chat Model.
Pitani Madhusayi
An open-source AI framework with Groq, FastAPI, RAG, ChromaDB and React SDK.
GitHub Actions
rulvar bridge adapter wrapping any Vercel AI SDK LanguageModelV4 as a ProviderAdapter.
scan5
AI security scanner and runtime enforcement for LLM applications — detects prompt injection, API key leaks, unsafe output, RAG poisoning, and agent hijacking in code and live traffic.
...moremarcosluizfp
looprun on the Vercel AI SDK — RESERVED (not implemented). The backend seam contract lives in @looprun-ai/core; @looprun-ai/mastra is the v0 backend.
...moreStewart Bonnick
OpenAI OAuth provider for AI SDK 7 and TanStack AI backed by the Codex Responses endpoint
Scott Nelson
Read-only Mac disk usage explorer with plain-English AI explanations and built-in safety knowledge. Scan, understand, reclaim.
...moreshirabe-dev
Official thin SDK for the Shirabe Japan-specific AI-native API platform — composite enrich (address, name, corporate number, calendar) in one call, plus ready-made Vercel AI SDK and LangChain tools. Zero runtime dependencies in the core.
...moretheabbie
bridge any raw llm stream to be used with pi
Wannes Vantorre
Caching based RAG primitives
GitHub Actions
Record automation run events in LumaTrack, the system of record for automation value.
Shun Li
Pi extension that syncs models from an OpenAI-compatible API and enriches them with pricing and capability metadata.
kitepon-rgb
UIに依存せず、ログイン済みChatGPT Web runtimeの通常ChatをCodexから呼び出すローカルconnector
stormie-bot
☢️ An AI SDK tool that allows AI models to leverage Power Plant generators to enhance their capabilities
edup
Docker Swarm CSI StorageDriver — pool = one cluster-scoped CSI volume; subPath = directory within. WS6 Part B of xema-os-plan §workspace.
...moreLaC (LLM as Code) - reference engine. Declarative behavior for LLM apps: law in files, perimeter in code.