According to AlphaMoat, as of 2026-09, MiroFish Predict has 1,489 downloads and 2 stars — ranked #4,775 of 5,078 AI skills overall, and #920 of 969 in AI Agent.
| Platform | Installs | Downloads | Stars | Comments |
|---|---|---|---|---|
| ClawHub ↗ | 31 | 1.5K | 1 | - |
| SkillHub ↗ | 141 | 1.5K | 2 | — |
Platforms count differently and update at different times.
It describes a workflow from entering a topic and building a knowledge graph to generating agents, running a simulation, and reviewing a report; follow-up questions are also supported. Results are simulated rather than factual forecasts or investment advice. Running it requires Docker and an LLM API key and may consume substantial model resources. The mirofish CLI and Docker backend are required; review container sources and monitor model costs.
Profile published 2026-10-29
Generated from the skill’s documentation as examples; not results of actual runs.
The counting method changed in 2026-09: skills with the same name are now counted separately by author, and the full catalog is covered. Dashed lines show months before the change; they are not compared with later months, and no growth is computed across the change.
| Month | Downloads | MoM | Stars |
|---|---|---|---|
| 2026-09 | 1.5K | — | 2 |
| 2026-08 | 1.3K | +18.7% | 2 |
| 2026-07 | 1.1K | +31.1% | 2 |
| 2026-06 | 838 | +18.2% | 2 |
| 2026-05 | 709 | — | 2 |
| 2026-03 | 336 | — | 1 |
The author’s original file, shown as plain text without edits or review by AlphaMoat. Links are not clickable; check commands before running them. Fetched 2026-10-23.
---
name: mirofish-predict
description: "MiroFish 群體智能推演引擎。當用戶要求「推演」「預測」「模擬」「如果…會怎樣」時使用。透過 55 個 AI Agent 在模擬社交平台上互動推演未來趨勢。"
homepage: https://github.com/666ghj/MiroFish
metadata:
{
"openclaw":
{
"emoji": "🐟",
"requires": { "bins": ["mirofish", "docker"], "env": ["LLM_API_KEY"] }, "primaryEnv": "LLM_API_KEY",
"install":
[
{
"id": "mirofish-npm",
"kind": "node",
"package": "mirofish-cli",
"bins": ["mirofish"],
"label": "Install MiroFish CLI (npm)",
},
],
},
}
---
# MiroFish 群體智能推演
用 55 個 AI Agent 在模擬的 Twitter/Reddit 上互動,推演未來趨勢。
## 一鍵推演
```bash
mirofish predict "如果比特幣突破15萬美元,加密市場會怎樣?"
```
自動完成:啟動 Docker 後端 → 建立專案 → 構建知識圖譜 → 生成 Agent → 運行模擬 → 輸出報告。
## 首次設定
需要 Docker Desktop 和 LLM API key。
```bash
# 1. 啟動(首次會自動拉 Docker image)
mirofish serve start
# → 如果沒有 ~/.mirofish/.env,會生成模板讓你填 API key
# 2. 配置 API key 後重新啟動
mirofish serve start
```
### API Key 設定方式
- 環境變數:`export LLM_API_KEY=xxx`
- OpenClaw config:`skills."mirofish-predict".env.LLM_API_KEY`
- 檔案:`~/.mirofish/.env`
## 命令參考
| 命令 | 功能 |
|:---|:---|
| `mirofish predict "主題"` | 完整推演(自動啟動後端) |
| `mirofish predict "主題" --rounds=10` | 指定推演輪數 |
| `mirofish serve start` | 啟動 Docker 後端 |
| `mirofish serve stop` | 停止後端 |
| `mirofish serve status` | 檢查後端狀態 |
| `mirofish projects` | 列出所有專案 |
| `mirofish status <sim_id>` | 查詢模擬進度 |
| `mirofish report <sim_id>` | 取得推演報告 |
| `mirofish chat <sim_id> "問題"` | 對報告追問 |
| `mirofish interview <sim_id> <agent_id> "問題"` | 採訪特定 Agent |
| `mirofish env` | 顯示設定狀態 |
## 追問
```bash
mirofish chat <sim_id> "哪些 KOL 的觀點最極端?"
mirofish interview <sim_id> 0 "你對 BTC 破 15 萬有什麼看法?"
```
## 注意事項
- 推演消耗大量 LLM token(55 Agent × N 輪),本地模型可節省費用
- `--rounds` 建議 10-20 輪先試,效果好再加到 40
- `--platform` 可選 `twitter`、`reddit`、`parallel`(推薦 parallel)
- 建議 LLM ≥ 14B 參數
Records learnings, errors and corrections as Markdown files for later reuse and promotion into workspace memory.
Finds skills in local and remote catalogs from a plain-language request and recommends what to install.
A browser-automation command line skill for AI agents.
Turns corrections and self-review into tiered memory and promotes or archives it by use.
Reviews a third-party skill's source, permissions, and danger patterns against a checklist before it is installed.
Stores agent memory and cross-skill state as a typed, constraint-checked knowledge graph.
A workspace setup guide for planning proactive follow-up, working memory and review routines for an AI agent.
A guide to creating and updating effective Claude skills.
Data month: 2026-09 · Downloads, stars and installs are aggregated monthly from public skill registries (ClawHub, SkillHub). See methodology.