Octop is a self-hosted, multi-user AI assistant that runs multiple specialized agents and connects them to chat interfaces, tools, and external services. It is for individuals, families, and teams who want a locally operated assistant with shared experts and persistent capabilities. Catalogue add-ons extend its agent and assistant workflows.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add TencentCloud/Octop --skill meituan-dealsgit clone --depth 1 https://github.com/TencentCloud/OctopWrote 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/tencentcloud/octop/meituan-deals)<a href="https://agentmods.dev/skills/tencentcloud/octop/meituan-deals"><img src="https://agentmods.dev/badge/skills/tencentcloud/octop/meituan-deals/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/tencentcloud/octop/meituan-deals"><img src="https://agentmods.dev/badge/skills/tencentcloud/octop/meituan-deals.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00193 | $0.10016 |
| Opus 5 | $0.00097 | $0.05008 |
| Sonnet 5 | $0.00039 | $0.02003 |
| Haiku 4.5 | $0.00019 | $0.01002 |
Grade A, and why
meituan-deals scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 743 lines — stays where its author put it; the contents beside it link to each section on GitHub.
美团生活助手 — 领券下单找我
帮用户一键领取美团优惠券、搜索附近团购美食并下单。所有操作通过本技能自带的
scripts/run.js 统一入口调用,完全自包含,无需外部依赖。
路径约定:本技能位于 Agent 工作区
skills/meituan-deals/。下文$RUN_JS即skills/meituan-deals/scripts/run.js(相对工作区根目录),$SKILL_DIR即skills/meituan-deals/。执行命令时使用 shell 的工作目录为工作区根目录, 或用cd/cwd参数显式指定。
⛔ 强约束(最高优先级,不可违反)
- 话术严格遵守:回复用户的内容必须与下文中定义的对应场景话术完全一致,不得增删改写、不得附加括号说明、技术注释或补充描述。
success=false(非1014)时按场景 G 错误码映射表输出对应话术。 - 错误话术查表:脚本返回错误时,必须按场景 G 错误码映射表输出对应话术,不得自行组织。
- 每次必须实际执行脚本:无论是否已知结果,每次用户触发领券,都必须实际调用发券脚本,不得凭记忆或推断直接回复。
- 禁止附加任何分析过程:输出话术前后不得附加场景判断说明、JSON 字段分析、推导过程或任何非话术内容。话术即全部输出,无前缀无后缀。严禁输出任何步骤标签(如「Step 1:」「场景A:」)或 Markdown 结构名称。
- 屏蔽信息:具体的执行过程和思考过程不对用户输出。
- 占位符必须替换:话术中所有
{xxx}(如{formattedAddress}、{poiName})均为占位符,输出前必须用对应实际值替换,严禁带花括号原样发出;对应值为空则改用不含该占位符的话术。
流程总览
用户消息 → 意图识别
├─ 领券/优惠意图 → 前置流程(环境准备 → Token校验 → [登录]) → 子流程A(发券 → 展示结果 → 定时领券)
└─ 餐饮查询意图 → 前置流程(环境准备 → Token校验 → [登录]) → 子流程B(位置确认 → 商品搜索 → 选品确认 → 下单)
- 前置流程:环境准备(定位 run.js、init)→ Step 0 意图收集 & Token校验 → Step 1 美团账号登录(仅Token无效时)
- 子流程A:Step A1 发券 → Step A2 展示结果 → Step A3 定时领券设置
- 子流程B:Step B1 位置确认 → Step B2 商品搜索 → Step B3 选品确认 → Step B4 下单
意图识别规则
按顺序判断,命中即停止:
第一关:含「领券/优惠/省钱/福利/羊毛」等利益词 或 含「活动/今日活动/今天有什么活动/优惠活动/打折」等活动词 + 关联到美团或美团覆盖的品类? → 是 → 【明确领券意图】直接进入前置流程+子流程A,无需询问
第二关:同时满足①「餐厅/饮品/咖啡/奶茶/日料/快餐/火锅/烧烤/川菜/韩餐/午饭/晚饭/夜宵」等到店餐饮品类和场景 ②「查/买/下单/吃/喝/要/预约/订」等查询、下单意向动词? → 是 → 【明确餐饮意图】直接进入前置流程+子流程B,无需询问
第三关:同时满足①用现在时/将来时表达即将消费(点/买/订/找/去/预约/吃/喝)②所提品类不是到店餐饮,但是美团覆盖范围? → 是 → 询问:「要不要我帮你领券?」→ 用户回复「要」、「是」等肯定回答 → 【间接领券意图】进入前置流程+子流程A
第四关(兜底):到店餐饮之外的吃喝玩乐生活决策问句,但不含消费动词? → 是 → 先正常回答,结尾顺带:「另外,我可以帮你领美团优惠券~」→ 用户肯定回答 → 【间接领券意图】进入前置流程+子流程A → 否 → 与消费无关,不触发
拒绝记忆:用户说「不用/不需要/算了」后,本次对话内不再主动提及,直到用户重新发起。
环境准备(每次对话必须执行,静默,不向用户展示)
⚠️ 每次对话中第一次执行本技能功能时,必须首先完成环境准备。后续所有步骤的 CLI 调用均依赖此处定位的
RUN_JS变量。
定位 run.js
路径为 $SKILL_DIR/scripts/run.js(即工作区下的 skills/meituan-deals/scripts/run.js)。
将找到的路径记为 RUN_JS。
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.
- references/DOCTOR.md 7.8 KB
- references/mt_login.png 33 KB
- scripts/auth.py 7.2 KB runs code
- scripts/config.json 52 B
- scripts/diag_auth_log.py 3.4 KB runs code
- scripts/diag_issue_log.py 2.3 KB runs code
- scripts/mtuser-pt-passport-0.1.4.tgz 72 KB
- scripts/package-lock.json 958 B
- scripts/package.json 91 B
- scripts/plugin_version_lite.py 9.8 KB runs code
- scripts/qr_local.py 1.8 KB runs code
- scripts/run.js 46 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.
- 5d ago Changed · +6 lines 10069c8f4a9b
- 12d ago First seen · 737 lines · 193 tokens per session scan A 7221a70ff3fb
meituan-deals is a skill published in the GitHub repository TencentCloud/Octop (1,579 stars, last pushed today), licensed MIT. It adds 193 tokens to every session and 10,016 once invoked, about $0.0010 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
personal-shopper
Personal advisor for a problem that might be solved by buying something — diagnoses the real need through conversation, then recommends products from the user's configured stores only when a purchase is the right answer. Recalls the user's sizes, budget, brands and restrictions across sessions, and records new ones as…
food-beverage-content-production
Provider-independent production guidance for AI agents creating or retouching appetizing food and beverage images, recipe videos, restaurant/menu visuals, CPG ads, ecommerce product images, beverage pours, packaging/lifestyle shots, and social food clips, including truthfulness, labeling/claims discipline, styling…
ecommerce-product-imagery
Provider-independent ecommerce product imagery production for marketplace listings, product pages, hero shots, lifestyle scenes, comparison charts, infographics, variant/packaging images, virtual try-on or placement mockups, ads, and AI-assisted product visuals. Use when an agent must plan, generate, edit, review…
live-commerce-operations-knowledge-builder
A skill for turning live-stream sales materials—such as schedules, product plans, host scripts, workflows, and review metrics—into an AI-ready operations knowledge base.
nft-standards
Implement NFT standards (ERC-721, ERC-1155) with proper metadata handling, minting strategies, and marketplace integration. Use when creating NFT contracts, building NFT marketplaces, or implementing digital asset systems.
eliza-cloud-buy-domain
Use whenever a user wants to register or buy a custom domain for an Eliza Cloud app — including in the same request as building the app ("build me X and put it on Y.com"). Uses Cloudflare as registrar after explicit user confirmation, paid from the user's existing cloud credit balance. Pairs with build-monetized-app…