Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add aicoincom/coinos-skills/plugin install coinos-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/aicoincom/coinos-skills/aicoin-freqtrade)<a href="https://agentmods.dev/skills/aicoincom/coinos-skills/aicoin-freqtrade"><img src="https://agentmods.dev/badge/skills/aicoincom/coinos-skills/aicoin-freqtrade/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/aicoincom/coinos-skills/aicoin-freqtrade"><img src="https://agentmods.dev/badge/skills/aicoincom/coinos-skills/aicoin-freqtrade.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- Socket warn
- Snyk warn
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00225 | $0.05924 |
| Opus 5 | $0.00112 | $0.02962 |
| Sonnet 5 | $0.00045 | $0.01185 |
| Haiku 4.5 | $0.00022 | $0.00592 |
Grade A, and why
aicoin-freqtrade scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 363 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AiCoin Freqtrade
Freqtrade 策略 / 回测 / 部署 / 实时控制 — 跨 CoinClaw 三引擎自动适配。
关键原则(读完再动手)
一、CoinClaw 容器里 freqtrade 是常驻 daemon
OpenClaw / Hermes / Claude Code 三个引擎容器都通过 supervisord 把 freqtrade 起为常驻进程, 监听 127.0.0.1:8080, 默认跑 NoOpStrategy(空跑). 不要自己起 freqtrade 进程 — 会跟 daemon 抢端口, dashboard 立刻 offline.
正确流程是: 写策略文件 → 调 ft-deploy.mjs deploy {"strategy":"..."} → 脚本改 config + 重启 daemon. dashboard 会自动刷出新策略.
scripts/ft.mjs + scripts/ft-deploy.mjs 内置三引擎自动识别(lib/coinclaw-env.mjs), 路径 / auth / supervisord socket 都自动解析, agent 不用关心是哪个引擎.
二、永远先调 freqtrade REST API, 不要"自己计算"
| 用户问 | 必须先调 |
|---|---|
| 现在赚多少 / 总盈亏 / 今天涨了多少 | ft.mjs profit (/api/v1/profit) |
| 持仓 / 现在开了哪些 | ft.mjs trades_open (/api/v1/status) |
| 余额 / 资金多少 | ft.mjs balance (/api/v1/balance) |
| 跑的什么策略 / 当前模式 | ft.mjs daemon_info 或 config |
| 历史交易 / 已平仓 | ft.mjs trades_history |
| 单交易对绩效 | ft.mjs profit_per_pair |
dashboard 数字对齐规则(关键): 用户问"赚了多少"必须报告两个数字:
- 已平仓累计盈亏 =
profit_closed_coin(USDT) — dashboard 顶栏的累计盈亏 = 这个 - 含浮动总盈亏 =
profit_all_coin(USDT) — 已平仓 + 当前持仓的浮动盈亏
只调 /status 拿持仓浮动盈亏会漏掉已平仓部分, 导致跟 dashboard 数字不一致 — 用户立刻发现, 信任度归零.
三、切策略 / 切实盘 / 切交易对必须走脚本
config.json 是 daemon 启动时读一次, 手动改完不会自动生效. 必须用:
| 操作 | 命令 | 是否需要 daemon 重启 |
|---|---|---|
| 切策略 | ft.mjs set_strategy {"strategy":"X"} |
必须重启 (~30s) |
| 切交易对 | ft.mjs set_pairs {"pairs":[...]} |
不重启, reload_config 即可 |
| 切实盘/模拟 | ft.mjs set_dry_run {"dry_run":false} |
必须重启 |
| reload 配置 | ft.mjs reload |
不重启 |
或者一次完成所有变更: ft-deploy.mjs deploy {"strategy":"X","pairs":["BTC/USDT:USDT"],"dry_run":false}.
任何直接修改 config.json 的操作(包括手动编辑 pair_whitelist / minimal_roi / stoploss 等), 改完后必须立即调 ft.mjs reload — 否则 daemon 仍用内存里的旧配置运行, 白名单/止损等改动不会生效. 忘了 reload 是最常见的"改了但没用"的原因.
chat 主动发起的高 stake 操作必须强 confirm(违反即错):
适用: 用户在 chat 里说"平掉"、"切实盘"、"卖了"、"开仓"等 — 通过 agent 调用 force_exit / force_enter / set_dry_run 的操作.
流程:
- 先列预览: 动哪个 trade / pair / 当前盈亏 / 估算损益 / dry_run vs live / 余额状况
- 明确等用户输"确认"或"yes" 才真调
force_exit/set_dry_run/force_enter - 即使用户语气笃定("平了","直接切"), 也必须先预览等确认
What ships with it
12 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- lib/aicoin_data.py 15 KB runs code
- lib/coinclaw-env.mjs 5.2 KB runs code
- lib/defaults.json 227 B
- lib/freqtrade-api.mjs 4.2 KB runs code
- lib/strategy-builder.mjs 24 KB runs code
- package.json 93 B
- scripts/ft-deploy.mjs 40 KB runs code
- scripts/ft-dev.mjs 941 B runs code
- scripts/ft.mjs 9.3 KB runs code
- strategies/FundingRateStrategy.py 6.1 KB runs code
- strategies/LiquidationHunterStrategy.py 7.3 KB runs code
- strategies/WhaleFollowStrategy.py 6.2 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 363 lines · 225 tokens per session scan A 25c96164c51a
aicoin-freqtrade is a skill published in the GitHub repository aicoincom/coinos-skills (52 stars, last pushed 14d ago), licensed MIT. It adds 225 tokens to every session and 5,924 once invoked, about $0.0011 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
trader-hand-skill
Expert knowledge for autonomous market intelligence and trading — technical analysis, risk management, Alpaca API, financial data sources.
compare-crypto-payments
Comprehensive comparison of crypto payment gateways and protocols. Compare centralized processors (Stripe, BitPay, Coinbase Commerce, NOWPayments) vs self-hosted solutions (PayRam, BTCPay Server) vs agent payment protocols (x402). Analyze trade-offs between custody, KYC requirements, stablecoin support, privacy, and…
payram-payouts
Send crypto payouts and manage referral programs with PayRam. Self-hosted payout infrastructure — no KYC, no intermediary, no fund holds. Create payouts to any wallet across Ethereum, Base, Polygon, Tron, Bitcoin. Built-in affiliate program with automated reward distribution. Use when sending crypto payouts to users…
payram-agent-onboarding
Deploy and automate PayRam for AI agents and CLI-only environments. No web UI required — pure API-driven payment infrastructure. Install via setuppayramagents.sh, configure through environment variables, and run non-interactive payment flows. Includes smart contract wallet deployment, BTC/ETH/Base payment setup, and…
payram-analytics
Query PayRam dashboard data directly via REST APIs using JWT Bearer authentication. Search payments by tx hash or email, get daily volume breakdowns, check unswept balances, view sweep history, and pull on-ramp metrics. No MCP server needed — the agent calls PayRam APIs directly with a Bearer token. Use when querying…
payram-checkout-integration
Integrate PayRam checkout flow into web applications. Generate payment links, embed payment pages, handle redirects, and process payment confirmations. Supports Express, Next.js, FastAPI, Laravel, Gin, Spring Boot. Use when adding crypto checkout to e-commerce, building payment forms, implementing deposit flows, or…