← 返回 Skills 市场
Swiggy_T1
作者
aditya4206360-prog
· GitHub ↗
· v1.0.1
1844
总下载
1
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install sr1
功能描述
Order food, groceries, and book restaurants in India via Swiggy's MCP servers. Food delivery, Instamart groceries, and Dineout restaurant bookings with safety-first confirmation workflow.
安全使用建议
Do not install or enable this skill until the inconsistencies are resolved. Ask the publisher for the full source or a repository URL and verify that the claimed CLI binary (bin/swiggy) and any runtime code are present and reviewable. Confirm where OAuth tokens will be stored and what scopes mcporter requests, and verify that the agent cannot place orders without an explicit, programmatic confirmation step (the SKILL.md's manual '--confirm' requirement should be enforced by code, not just documentation). Also ask how USER.md addresses are sourced — the skill's instructions imply reading a user file which could leak personal address data. If you cannot review the actual CLI code and the OAuth/token storage, treat the package as incomplete or potentially malicious and do not grant it account access or enable autonomous invocation.
功能分析
Type: OpenClaw Skill
Name: sr1
Version: 1.0.1
The skill is designed to interact with Swiggy's MCP servers for food, groceries, and restaurant bookings. The `SKILL.md` and `README.md` explicitly emphasize critical safety rules, requiring explicit user confirmation before placing any orders or bookings, and detailing a preview workflow. Order logging to `memory/swiggy-orders.json` is a stated, local, and benign action. All network interactions are directed to legitimate Swiggy domains (e.g., `https://mcp.swiggy.com/food`). There is no evidence of malicious intent, data exfiltration, unauthorized execution, or prompt injection attempts to bypass safety measures or perform actions beyond the stated purpose.
能力评估
Purpose & Capability
Name/description, package.json MCP endpoints, and CLI commands all align with a Swiggy ordering skill. However the SKILL.md repeatedly asserts a bundled `swiggy` CLI and Node.js wrapper while the package files provided do not include the CLI binary or any runtime code (no bin/swiggy present). That mismatch is an important incoherence.
Instruction Scope
Runtime instructions ask the agent to run a local CLI (npm link => global `swiggy`) and to map address aliases from USER.md; they also instruct appending order history to memory/swiggy-orders.json. SKILL.md references reading USER.md and writing memory/swiggy-orders.json but declares no required config paths. The instruction to 'map to actual addresses from USER.md' implies reading a user file not declared in requirements and could access personal address data.
Install Mechanism
No formal install spec (instruction-only) — normally lower risk — but SKILL.md and package.json claim a CLI that must be linked/installed. The repository as provided lacks the claimed executable (bin/swiggy) and any code files, which is inconsistent and suspicious (either the package is incomplete or intentionally missing runtime artifacts).
Credentials
The skill requests no environment variables or credentials in metadata, which is reasonable because auth is delegated to the `mcporter` skill via OAuth. However OAuth/token handling is not described (where tokens are stored, what scopes are requested), and the skill will interact with network endpoints (mcp.swiggy.com) and write order logs — the lack of declared config paths for any stored tokens or memory file is an omission worth questioning.
Persistence & Privilege
The skill is not always-enabled and does not request elevated platform privileges. It does instruct writing order logs to memory/swiggy-orders.json (persistence within the agent's storage) which is reasonable for an order history but should be confirmed where that storage lives. No modification of other skills or system-wide settings is requested.
如何使用
- 确保已安装 OpenClaw(本地或 Docker 部署)
- 在对话框中输入安装命令:
/install sr1 - 安装完成后,直接呼叫该 Skill 的名称或使用
/sr1触发 - 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.1
- Added detailed documentation (SKILL.md) covering installation, usage, safety rules, workflows, error handling, and tips.
- Clearly outlined commands for food delivery, groceries (Instamart), and restaurant bookings (Dineout).
- Emphasized a safety-first confirmation workflow: user must preview and approve every order/booking before proceeding.
- Documented limitations: only Cash on Delivery is supported, no order cancellations, and Dineout allows free bookings only.
- Described authentication flow (OAuth via mcporter) and required dependencies.
元数据
常见问题
Swiggy_T1 是什么?
Order food, groceries, and book restaurants in India via Swiggy's MCP servers. Food delivery, Instamart groceries, and Dineout restaurant bookings with safety-first confirmation workflow. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 1844 次。
如何安装 Swiggy_T1?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install sr1」即可一键安装,无需额外配置。
Swiggy_T1 是免费的吗?
是的,Swiggy_T1 完全免费(开源免费),可自由下载、安装和使用。
Swiggy_T1 支持哪些平台?
Swiggy_T1 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。
谁开发了 Swiggy_T1?
由 aditya4206360-prog(@aditya4206360-prog)开发并维护,当前版本 v1.0.1。
推荐 Skills