himorishige avatar

team-ai-kb

v0.2.0

by himorishige

Agent-facing layer of the team AI environment: knowledge base search (rag-kb + nvidia-rag MCP), web search (web-search skill), and Backlog operations via bee CLI (using-bee skill)

What team-ai-kb does, in the publisher's words

A local router that puts NeMo Switchyard between opencode and Fireworks AI, automatically dispatching each request to a strong or weak tier so you cut cost without giving up quality. It runs as a Docker container on your own machine.

Everything is configured in a single file, routes.toml (the config format of Switchyard's native Rust server).

Attaching images: use k3-only. It pins Kimi K3, which reads images, and it is the only route that declares modalities in the opencode config so the attachment is actually sent, opencode silently drops an image for any model that does not declare it, and the model then replies that it cannot read images.

From the project README, punctuation lightly normalized · Open the README on GitHub

Skills
3
MCP servers
0
Stars
2
License
Apache-2.0
Repo created
2026-07-23
Last pushed
2026-09-02
Publisher type
User
Version
0.2.0

Links

Skills · 3

rag-kb
Guide to searching the shared team knowledge base (NVIDIA RAG Blueprint) over MCP. Use when: you need something from internal knowledge, verification records or published articles / the user asks you to "check the knowledge base" or "search the RAG" / you need to confirm a number from a past experi…
using-bee
Use when interacting with Backlog project management service via the bee CLI: creating or editing issues, listing pull requests, managing projects/wikis/documents, checking notifications, or opening Backlog pages.
web-search
Grounded web search through Gemini (Google Search grounding) or OpenAI (web search). Returns an answer with its sources. Use when: you need current or public information from the web / the user asks you to "search the web" or "google it" / you need general knowledge, a recent release, a price or a…

Descriptions come from the frontmatter of each SKILL.md, punctuation lightly normalized.

Category

Research & Search. Assigned by this directory. Agent Plugins 1.0.0 has no category field, so no manifest declares one.

Related by publisher and keywords

  • World Monitor agent plugin: live global-intelligence MCP servers plus task-focused agent skills for country risk, conflict, markets, chokepoints, energy, climate, and related situational awareness.

  • mem0mem0ai

    Cross-session memory and token savings for coding agents.

  • context7upstash

    Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.

  • hyperframesheygen-com

    HyperFrames by HeyGen. Write HTML, render video. Compositions, GSAP and runtime adapter animations, captions, voiceovers, audio-reactive visuals, and website capture for HyperFrames.

  • chrome-devtoolsChromeDevTools

    Reliable Chrome automation, debugging, and performance analysis for AI agents.

  • Portable skills-only "AAS Accessibility & Inclusive UX" plugin from Agentic Awesome Skills.

Add the verified badge to your README[![Verified by AgentPluginsDirectory.com against Schema 1.0.0](https://agentpluginsdirectory.com/badge.svg)](https://agentpluginsdirectory.com/plugins/team-ai-kb)

✓ Verified . We fetched the manifest from GitHub and checked it against the official Agent Plugins 1.0.0 schema at agent-plugins.org.