npm Registry (Real-time)
ActiveReal-time npm registry change stream for AI/MCP packages
38,834
Items Listed
N/A
Sync Frequency
api
Access Method
9/26/2026
Last Synced
Items in this Directory
ls-how-to-find-balanced-coverage-of-a-topic-a-quick-guide-mtmnuls0
linkstack
How to find balanced coverage of a topic — a quick guide
oai-tts-mcp
Komorebi
OpenAI-compatible TTS MCP server with voice discovery and public audio URLs.
@tweaktags/storage-s3
Scarlett A. Scott (codescarlett)
S3 and S3 compatible media storage adapter for TweakTags, using presigned uploads
@go-hare/claude-code-darwin-x64
go-hare
Platform-specific binary for @go-hare/claude-code (macOS x64)
curatively-mcp
abaii
MCP server for Curatively — search, read and save to your private vault from Claude Code, Codex, Cursor or any MCP client.
...moreagentic-e2e
dhevenddra
Plain-English end-to-end tests run by an AI agent against an ephemeral preview of your PR's build. Self-healing, with a video replay of every failure. Built on Solari.
...more@agentkit-js/tools-rag
agentkit-js contributors
RAG / embedding tools for agentkit-js — ModelEmbedder + ragTool + Pinecone/Qdrant/Local connectors
@hasna-internal/kai-agent-default-model
andreihasna2
Default model selection shared by Agent entry points
skill-mcp-manager.skill
smm contributors
Unified skill & MCP manager for Claude Code — 11 slash commands
@hasna-internal/kai-agent-instructions
andreihasna2
Workspace context loader for AGENTS.md/CLAUDE.md instruction files
@hasna-internal/kai-agent-loop
andreihasna2
The concrete agent loop plugin for the DeepSeek Harness
@hasna-internal/kai-agent-loop-testkit
andreihasna2
Shared prerequisite mounting for tests that exercise the concrete agent loop
@wasmagent/mcp-gateway
tellerlin
MCP Gateway — identity propagation, server card validation, policy enforcement, AEP evidence emission for MCP tool invocations
...more@hasna-internal/kai-agent-presets
andreihasna2
Per-session agent composition from preset cordis.yml files for the DeepSeek Harness
@hasna-internal/kai-agent-spine-demo
andreihasna2
The default executor-less/UI-less agent spine with fallback session titles, provider-routed retry, and optional persisted goals
...more@agenticprimitives/intent-marketplace
richcanvas
Direct Lane intent matchmaking. Intent + ConstraintSet + AssumptionSet + ResolutionReceipt + IntentMatch + composite-score ranking. Resolver layer 4 folded into PassThrough stub in W1.
...morenodeblink-mcp
NodeBlink Team
NodeBlink MCP server — lets AI agents create and verify USDC settlements on Solana (non-custodial, on-chain verified)
@hasna-internal/kai-agent-tool-presentation
andreihasna2
Agent-plane presentation selector: composes one agent's tools as Code Mode, native, or both
@davepi/mcp
davepi
One-line MCP wiring for dAvePi. Connects Claude Desktop / Cursor / Claude Code to a dAvePi instance — either a remote HTTP /mcp endpoint or a local stdio session.
...more@hasna-internal/kai-hooks-claude-code
andreihasna2
Bridge plugin: run a Claude Code hooks.json / settings hook config on the DeepSeek Harness interception seams