agentdb-advanced
Master advanced AgentDB features including QUIC synchronization, multi-database management, custom distance metrics, hybrid…
Route agentic-flow work across Anthropic, OpenRouter, Gemini and local ONNX models, and run the Anthropic-compatible translation proxy with its per-run token. Use when the user wants cheaper models, a non-Anthropic provider, local inference, or to point Claude Code at OpenRouter
Claude Code
$ npx -y skills add ruvnet/agentic-flow --skill multi-provider-routing --agent claude-code
Codex
$ npx skills add ruvnet/agentic-flow --skill multi-provider-routing -a codex -y
Cursor
$ npx skills add ruvnet/agentic-flow --skill multi-provider-routing -a cursor -y
Gemini CLI
$ npx skills add ruvnet/agentic-flow --skill multi-provider-routing -a gemini-cli -y
GitHub Copilot
$ npx skills add ruvnet/agentic-flow --skill multi-provider-routing -a github-copilot -y
OpenCode
$ npx skills add ruvnet/agentic-flow --skill multi-provider-routing -a opencode -y
Windsurf
$ npx skills add ruvnet/agentic-flow --skill multi-provider-routing -a windsurf -y
name: multi-provider-routing description: Route agentic-flow work across Anthropic, OpenRouter, Gemini and local ONNX models, and run the Anthropic-compatible translation proxy with its per-run token. Use when the user wants cheaper models, a non-Anthropic provider, local inference, or to point Claude Code at OpenRouter or Gemini.
Set these with `/plugin` (configure) or in settings under `pluginConfigs`:
A key set in the plugin **replaces** the same variable from your shell for the MCP server. Left empty, the server sees an empty value, not your shell's: set the key in the plugin if you use that provider through it. `ANTHROPIC_API_KEY` is not a plugin setting; the server inherits it, or your Claude login, from the environment.
Never pass a key as a tool argument (`anthropicApiKey`, `openrouterApiKey`): arguments are written to the transcript, and the plugin's guard refuses such calls.
`agentic_flow_agent` (registered under the `learning` or `admin` profile) takes `provider` and `model` per call, overriding the defaults. Pick with the `model-router` agent when unsure.
agentic-flow can serve an Anthropic-compatible endpoint backed by OpenRouter or Gemini. Since 3.0.0 the proxy:
# terminal 1: the provider key stays here, in the proxy's environment OPENROUTER_API_KEY=... npx agentic-flow proxy --provider openrouter --model deepseek/deepseek-chat # terminal 2: use the token the proxy printed export ANTHROPIC_BASE_URL=http://127.0.0.1:3000 export ANTHROPIC_API_KEY=<token printed by the proxy> claude
Do not put a placeholder or dummy value in `ANTHROPIC_API_KEY`: the proxy rejects anything but its token. Do not bind the proxy to `0.0.0.0` on a shared network: whoever reaches the port and has the token spends your provider key.
`provider: onnx` runs on this machine with no key and no network for inference. Expect lower quality; use it for mechanical tasks or private text.
Production-ready AI agent orchestration with 66 self-learning agents, 213 MCP tools, and autonomous multi-agent swarms.
Master advanced AgentDB features including QUIC synchronization, multi-database management, custom distance metrics, hybrid…
Create and train AI learning plugins with AgentDB's 9 reinforcement learning algorithms. Includes Decision Transformer,…
Implement persistent memory patterns for AI agents using AgentDB. Includes session memory, long-term storage, pattern learning,…
Optimize AgentDB performance with quantization (4-32x memory reduction), HNSW indexing (150x faster search), caching, and batch…
Implement semantic vector search with AgentDB for intelligent document retrieval, similarity matching, and context-aware…
Quantum-resistant, self-learning version control for AI agents with ReasoningBank intelligence and multi-agent coordination
You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or…
Ask which skill or flow fits your situation. A router over the skills in this repo.