title	emoji	colorFrom	colorTo	sdk	pinned	hf_oauth	hf_oauth_scopes	license	short_description
cluas_huginn	💬	yellow	purple	docker	false	true	inference-api	apache-2.0	A gathering of guides, a council of counsels

🐦‍⬛ cluas huginn (thought's ear) 🐦‍⬛

- A Multi-Agent Research Council

CLUAS HUGINN

A Multi-Agent Deliberation Engine

Anno MMXXV — MCP 1st Birthday Hackathon Edition

<style> .cluas-banner { width: 100%; display: flex; justify-content: center; margin: 24px 0 32px; } .cluas-banner-inner { background: #f5f4ef url('/file=static/paper.png') repeat; background-size: 300px; border: 2px solid rgba(139, 88, 40, 0.55); /* copper ink */ padding: 24px 32px; border-radius: 12px; max-width: 720px; text-align: center; font-family: Labrada, serif; box-shadow: 0 0 1px rgba(0,0,0,0.05), 0 2px 6px rgba(0,0,0,0.06), 0 6px 12px rgba(0,0,0,0.04); } .cluas-banner-inner h1, .cluas-banner-inner h2 { text-shadow: 0 1px 2px rgba(0,0,0,0.1); } /* Title */ .cluas-banner-inner h1 { font-size: 1.9rem; margin: 0; color: #4a3524; /* ink brown */ font-weight: 700; letter-spacing: 0.02em; } /* Subtitle */ .cluas-banner-inner h2 { font-size: 1.05rem; margin: 6px 0 10px; font-weight: 500; color: #6a5648; letter-spacing: 0.03em; } /* Meta line */ .cluas-banner-meta { font-size: 0.9rem; color: rgba(70, 50, 35, 0.75); font-style: italic; letter-spacing: 0.04em; } </style>

A gathering of guides, a council of counsels

What It Does

Four specialised AI agents with distinct epistemic roles debate questions using structured dialectic, building knowledge over time.

• Thesis: Present findings with evidence

• Antithesis: Challenge, verify, and provide counterpoints

• Synthesis: Build consensus and update shared memory

• Evolution: Future discussions build on accumulated knowledge

Key Innovations / What Makes This Different

1. Unified inheritance architecture underpinning multi-agent system
2. Shared epistemic principles but distinct epistemic roles for each agent
3. "Four Temperament" character differentiation
4. Tool-use heuristics per character, guided by each character’s personality
5. Steelmanning and collaborative disagreement built-in
6. Structured dialectic (thesis → antithesis → synthesis) to reduce bias
7. Persistent memory enabling knowledge accumulation over time

Modes

• Collaborative Mode: Ask a question and receive synthesised research with sources
• Active Mode: Join the discussion, steer research, challenge claims, be 'the human in the loop'

The Council

Agent	Role	Icon
Corvus	Academic verifier (literature)	🐦‍⬛
Raven	Accountability enforcer (news, verification)	🦅
Magpie	Trend explorer (patterns, connections)	🪶
Crow	Grounded observer (data, environment)	🕊️

Dialectic Process

1. Thesis: Characters present initial findings using specialized tools
2. Antithesis: Characters challenge, verify, and provide counter-evidence
3. Synthesis: Council builds consensus and adds to collective memory
4. Evolution: Future discussions build on accumulated knowledge

Why It Matters

Most AI assistants are stateless. cluas_huginn remembers, learns, and builds knowledge over time.

🛠 Tech Stack

• Base: Python, Gradio
• LLMs: UserKeys/Groq/Nebius/etc (various models, llama3.3 main)
• Tools: Academic search, news verification, web exploration
• Memory: Persistent character memories

Taglines

• "A gathering of guides, a council of counsels"
• "Research that remembers, knowledge that accumulates"
• "Multi-agent MCP research collective"

On linked in:

https://www.linkedin.com/posts/james-boag_cluas-huginn-is-almost-ready-to-submit-activity-7401042694279467008-9eKI

on youtube:

https://youtube.com/playlist?list=PLTwFAkhmI_YV0U_f9csf7mdtkhFdhEo5c&si=NtNSIXyJ-j3q389G

This project is licensed under the Apache 2.0 License.