communication MCP Server
Destiny Atelier — single-file local AI chat. Three brains vote on every answer. 540 KB. 108 MCP tools. Open-source local Claude / GPT alternative. Local LLM. Runs on your own hardware.
Discovered via github-topic:mcp-server and last synced 3mo ago.
1. Install the package
npm install nandai-chat
Role
**Instruction-following**
`add_observations`, `create_entities`, `search_nodes`
**Factuality**
Examples
`read-note`, `search-vault`, `create-note`
llama.cpp-CPU
Notes
✅
Port
_TBD_
vLLM
443
—
What
Yours (proven on 2× RTX 3090 — 24 GB + 24 GB)
The whole stack is one HTML file
RTX 3090 Ti @ vLLM
RTX 3090 @ llama.cpp-CUDA
CPU @ llama.cpp
.213 :8056 (FastAPI)
Anthropic API
Failure mode it catches
**Tool-correctness**
**Refusal-appropriateness**
**Hallucination**
**Style-fit**
**Safety**
**Operator-cost**
`sequentialthinking` (multi-step CoT)
`create_issue`, `get_pull_request`, `search_repositories`
`search`, `fetch_content`
`fetch` (any URL)
`scrape`, `crawl`, `search`
`read_file`, `write_file`, `directory_tree`
`chat`, `business_snapshot`, `analyze_review`
`get_products`, `get_orders`, `get_inventory`
`system_health`, `gpu_status`, `ssh_exec`
`get_current_time`, `convert_time`
3057
91.4%
Fork it, white-label it, ship it.
llama.cpp-CUDA
Our model
Open WebUI
Reproducibility commitment: every benchmark row gets a `bench/<name>/` dir with the exact harness, the exact prompts, the model versions, the raw output, and a one-command rerun. No paper-quoted figures presented as our own. --- ## Build journal — one night, 18 ticks This chat was built across 18 chronicled "ticks" in one overnight session (May 15 → May 16, 2026). Each tick: one feature, one fix, one ship, one verify. The full chronicle: [`journal/`](./journal/) — 35 KB of build narrative + 30+ verification screenshots. A representative tick: > [HANDOFF-009](./journal/HANDOFF-009.md) — Sentinel wired > > **Attempted:** post-turn audit of every assistant reply against a 5-axis rubric. > > **Changed:** new `src/lib/sentinel.ts` (149 lines). New API call to > `/sentinel/verdict` endpoint on .213:8056. New `AuditPill` component > rendered below assistant turns. > > **Verified:** sent the canonical "what time is it in Tokyo right now?" > Sentinel returned `factuality: 92, instruction-following: 88, ...` within > 4.1 s. Verdict pill rendered in green. Re-asked with a deliberately > garbled tool spec; Sentinel flagged it amber and suggested > `rerun_with_tools`. One click reran successfully. > > **Broke:** AuditPill double-fired on browser back. Fixed in tick-010 by > tagging `reranWithTools: true` on the message after a rerun. Read the [whole journal](./journal/) start-to-finish; it's the best documentation of how the pieces compose. --- ## Comparison
Send messages and upload files to Slack channels using the `slack` CLI. Supports direct messages, channel posts, file uploads, and thread replies.
Bidirectional Telegram bridge — talk to Claude from your phone. Built into OpenPaw.
Discover agents across local node and mesh peers at runtime — their roles, handles, and how to communicate with or delegate to them.
Apply data visualization and storytelling principles from Storytelling with Data by Cole Nussbaumer Knaflic. Covers choosing effective visuals (line, bar, table, scatterplot), eliminating clutter (Gestalt principles, data-ink ratio), focusing attention (preattentive attributes, strategic color), thinking like a designer (affordances, alignment, white space), and narrative structure (three-act story, horizontal/vertical logic, Big Idea). Trigger on "data visualization", "chart design", "dashboard design", "data storytelling", "presentation chart", "declutter chart", "bar chart", "line chart", "data narrative", "slide deck data", "chart review", "viz critique", "storytelling with data".
🔬 A curated collection of 23,000+ agent skills for empirical research across 8 social science disciplines. | 精选 23,000+ AI Agent 技能库，覆盖8大社会科学学科的实证研究。CoPaper.AI 20分钟完成一篇可复现的规范实证论文，并支持用户上传 Skills。-- Maintained by CoPaper.AI from Stanford REAP.
🔬 A curated collection of 23,000+ agent skills for empirical research across 8 social science disciplines. | 精选 23,000+ AI Agent 技能库，覆盖8大社会科学学科的实证研究。CoPaper.AI 20分钟完成一篇可复现的规范实证论文，并支持用户上传 Skills。-- Maintained by CoPaper.AI from Stanford REAP.
AgentNetworkProtocol(ANP) is an open source protocol for agent communication. Our vision is to define how agents connect with each other, building an open, secure, and efficient collaboration network for billions of intelligent agents.
Learn how to use the c-slack Claude skill. Complete guide with installation instructions and examples.
Learn how to use the c-telegram Claude skill. Complete guide with installation instructions and examples.
Learn how to use the Letter Writer Claude skill. Complete guide with installation instructions and examples.