polymarket-telegram-picks

每日拉取 Polymarket 体育/赛事赔率，经 AI 分析后推送值得下注的推荐到 Telegram

Security Scan

The skill's code and instructions match its stated purpose (fetch Polymarket odds and send Telegram messages) but the package metadata omits required environment variables and there are small inconsistencies (unused/optional env checks) that should be clarified before installation.

Purpose & Capability

The name/description (fetch Polymarket NBA odds, AI analyze, push to Telegram) match the included scripts: fetch_polymarket.py (fetch), send_telegram.py (push), run_daily.py (orchestration). However the skill registry metadata declares no required env vars/credentials while the code and SKILL.md clearly require TELEGRAM_BOT_TOKEN and TELEGRAM_CHAT_ID — this is an inconsistency in declared requirements.

Instruction Scope

SKILL.md instructs the agent to run the included fetch and send scripts and to perform AI analysis based on the script output. The scripts themselves only fetch public Polymarket API data and call Telegram's Bot API; they do not read unrelated system files or attempt to transmit data to unexpected endpoints. The SKILL.md and scripts reference the Telegram credentials (env or config) which is within the scope of the described functionality.

Install Mechanism

This is an instruction-and-script-only skill with no install spec. No external downloads or package installs are requested. Scripts use only Python standard library (urllib, json, subprocess), so installation risk is low.

Credentials

The code legitimately needs TELEGRAM_BOT_TOKEN and TELEGRAM_CHAT_ID to send messages, and SKILL.md documents that. But the registry/metadata lists no required env vars — a mismatched declaration. run_daily.py also looks for OPENAI_API_KEY and POLYMARKET_DAILY_RAW_ONLY environment variables (OPENAI_API_KEY controls an optional local LLM path, POLYMARKET_DAILY_RAW_ONLY toggles behavior). OPENAI_API_KEY is not required for normal operation and no built-in LLM call is implemented in the current scripts, but the presence of these checks is unexpected and should be clarified in metadata and docs before trusting environment config.

Persistence & Privilege

The skill does not request 'always: true' and does not attempt to modify other skills or system-wide settings. It runs only when invoked (user-invocable or via cron) and its scripts operate within the skill directory.

What to consider before installing

Key points before installing: - Metadata mismatch: the skill package metadata declares no required env vars, but SKILL.md and scripts require TELEGRAM_BOT_TOKEN and TELEGRAM_CHAT_ID (or a config/config.json). Ensure you provide these securely (prefer environment variables) and do not commit them to source control. - Code behavior: the scripts only fetch public Polymarket data and send messages to Telegram using your bot token. They do not exfiltrate other files or credentials. Still, review the code yourself to confirm it matches your expectations. - Optional env vars: run_daily.py checks OPENAI_API_KEY and POLYMARKET_DAILY_RAW_ONLY. OPENAI_API_KEY is not used by the included code currently (no LLM call is implemented), but could be referenced by future modifications—be cautious about placing sensitive API keys in an environment where third-party skills run. - Operational recommendations: run in an isolated environment or sandbox, set Telegram credentials as environment variables rather than config files, and avoid committing config/config.json to version control. Consider the legal/ethical implications of automated betting recommendations in your jurisdiction. - If you want to proceed: update the skill metadata to declare required env vars, or at minimum verify and store TELEGRAM_BOT_TOKEN and TELEGRAM_CHAT_ID securely; if you need strict assurance, run the scripts manually first to observe network calls and outputs. Confidence note: high — the code is straightforward and consistent with the described function, but the omitted required env vars and the presence of optional env checks are clear inconsistencies that justify a 'suspicious' label until clarified.

Polymarket 每日赔率分析与 Telegram 推送

功能说明

当用户或定时任务触发「今日 Polymarket 分析」「Polymarket 赔率推送」「执行每日推荐」等请求时，按以下流程执行：

1. 拉取当日未开赛 NBA 赔率：运行本技能目录下的 scripts/fetch_polymarket.py，获取当日（中国时间）、且尚未开赛的 Polymarket NBA 比赛及赔率数据。
2. AI 分析：根据脚本输出的结构化数据，结合你自己的判断（赔率价值、热度、流动性等），筛选出你认为值得下注的球队/选项，并简要说明理由。
3. 推送到 Telegram：将你的分析结论（推荐列表 + 理由）通过 scripts/send_telegram.py 发送到用户配置的 Telegram 对话。

执行步骤

1. 在技能目录下执行：

   python3 scripts/fetch_polymarket.py
   

   或使用工作区中本技能路径下的 scripts/fetch_polymarket.py。脚本会输出当日、未开赛的 NBA 比赛摘要（事件标题、开赛时间、选项与隐含概率等）。

2. 阅读脚本输出，从中挑选你认为有下注价值的项目（例如某方赔率被低估、或与你的判断存在价值差）。用简洁中文写出推荐列表与理由。

3. 将上述分析结论作为内容，调用推送脚本：

   python3 scripts/send_telegram.py "你的分析内容（推荐赛事 + 理由）"
   

   或通过 stdin 传入：

   echo "你的分析内容" | python3 scripts/send_telegram.py
   

环境与配置

• Telegram 推送依赖环境变量（或在 config/config.json 中配置）：
  • TELEGRAM_BOT_TOKEN：Bot 的 token（由 @BotFather 获取）
  • TELEGRAM_CHAT_ID：接收消息的聊天 ID（私聊或群组）
• 若未配置，send_telegram.py 会报错并提示配置方式；分析步骤仍可执行，仅推送会失败。

定时任务（每日 8 点中国时间）

在 OpenClaw 中可使用 cron 在每天北京时间 8:00 触发一次本流程，例如：

openclaw cron add --name "Polymarket每日推送" \
  --cron "0 8 * * *" \
  --session main \
  --message "请执行今日 Polymarket 赔率分析并推送到 Telegram"

若使用系统 crontab（不经过 OpenClaw 会话），可配置为 8 点运行 scripts/run_daily.py（该脚本会拉取数据、可选调用本地/远程 LLM 分析、再推送 Telegram），详见 README。

输出格式建议

推送至 Telegram 的内容建议包含：

• 日期
• 今日推荐项（赛事/问题 + 推荐选项 + 简要理由）
• 可选：风险提示（预测市场有风险等）

分析结论完全由你（AI）基于当前赔率与上下文自行判断，无需与任何外部投注建议绑定。