Summoner agents

🚀 Summoner Agents Library Levels Up: Knowledge Connectors, GPT + MCP Variants

Summoner’s agents library now goes beyond coordination templates and into end-to-end agent workflows: agents that can pull from real knowledge sources, route through multiple integration styles, and turn retrieved context into structured outputs.

agents/
├── agent_ChatAgent_0..3
├── agent_ConnectAgent_0..1
├── agent_EchoAgent_0..2
├── agent_ExamAgent_0..1
├── agent_GameMasterAgent_0..1
├── agent_GamePlayerAgent_0..2
├── agent_GPTArXivAgent           <-- 🆕 GPT-guarded, optional direct API calls (ArXiv)
├── agent_GPTClusterAgent
├── agent_GPTGitHubAgent          <-- 🆕 GPT-guarded, optional direct API calls (GitHub)
├── agent_GPTNotionAgent          <-- 🆕 GPT-guarded, optional direct API calls (Notion)
├── agent_GPTPubMedAgent          <-- 🆕 GPT-guarded, optional direct API calls (PubMed)
├── agent_GPTRedditAgent          <-- 🆕 GPT-guarded, optional direct API calls (Reddit)
├── agent_GPTRespondAgent
├── agent_GPTSlackAgent           <-- 🆕 GPT-guarded, optional direct API calls (Slack)
├── agent_GPTWikipediaAgent       <-- 🆕 GPT-guarded, optional direct API calls (Wikipedia)
├── agent_HSAgent_0..1
├── agent_HSBuyAgent_0
├── agent_HSSellAgent_0
├── agent_InputAgent
├── agent_LangChainCrewAIAgent
├── agent_MCPArXivAgent           <-- 🆕 GPT-guarded, optional MCP server calls (ArXiv)
├── agent_MCPGitHubAgent          <-- 🆕 GPT-guarded, optional MCP server calls (GitHub)
├── agent_MCPNotionAgent          <-- 🆕 GPT-guarded, optional MCP server calls (Notion)
├── agent_MCPPubMedAgent          <-- 🆕 GPT-guarded, optional MCP server calls (PubMed)
├── agent_MCPRedditAgent          <-- 🆕 GPT-guarded, optional MCP server calls (Reddit)
├── agent_MCPWikipediaAgent       <-- 🆕 GPT-guarded, optional MCP server calls (Wikipedia)
├── agent_OrchBridgeAgent
├── agent_RateLimitAgent_0..2
├── agent_RecvAgent_0..2
├── agent_ReportAgent_0..1
├── agent_SendAgent_0..1

🧩 What’s New

• A full knowledge-source lineup: native agents for ArXiv, PubMed, Wikipedia, Reddit, GitHub, Notion, and Slack.

• Two integration tracks side-by-side:

  • GPT* agents: GPT-guarded agents that process an input and optionally convert it into direct API calls.
  • MCP* agents: GPT-guarded agents that process an input and optionally convert it into calls to an MCP server for the same source.

---

Want to try the repo? Set up the SDK and install agent deps (more here):

source build_sdk.sh setup && bash install_requirements.sh

Then pick an agent folder and run it (examples + agent list): repo README.

---

Full Changelog: v1.0.0...v1.1.0

Summoner agents

🚀 Summoner Agents Library Keeps Expanding!

Summoner now includes 30+ native agents — all ready for high-throughput async coordination and multi-agent experimentation.

agents/
├── agent_ChatAgent_0
├── agent_ChatAgent_1
├── agent_ChatAgent_2
├── agent_ChatAgent_3
├── agent_ConnectAgent_0
├── agent_ConnectAgent_1
├── agent_EchoAgent_0
├── agent_EchoAgent_1
├── agent_EchoAgent_2
├── agent_ExamAgent_0
├── agent_ExamAgent_1
├── agent_GameMasterAgent_0
├── agent_GameMasterAgent_1
├── agent_GamePlayerAgent_0
├── agent_GamePlayerAgent_1
├── agent_GamePlayerAgent_2
├── agent_GPTClusterAgent
├── agent_GPTRespondAgent
├── agent_HSAgent_0
├── agent_HSAgent_1
├── agent_HSBuyAgent_0
├── agent_HSSellAgent_0
├── agent_InputAgent
├── agent_LangChainCrewAIAgent   <-- 🆕 native CrewAI + LangChain integration
├── agent_OrchBridgeAgent        <-- 🆕 general orchestration bridge (drop-in for CrewAI, LangChain, etc.)
├── agent_RateLimitAgent_0
├── agent_RateLimitAgent_1
├── agent_RateLimitAgent_2
├── agent_RecvAgent_0
├── agent_RecvAgent_1
├── agent_RecvAgent_2
├── agent_ReportAgent_0
├── agent_ReportAgent_1
├── agent_SendAgent_0
├── agent_SendAgent_1

🧩 What’s New

• LangChainCrewAIAgent – native integration to run CrewAI or LangChain-powered agents in Summoner
• OrchBridgeAgent – refactored version designed as a drop-in adapter for agents built in other orchestration frameworks
• Async performance improvements for multi-agent communication
• Expanded agent templates for simulation and evaluation challenges