印度点餐助手
india-food-ordering
by anugotta
Unified food ordering assistant for India that supports Swiggy and Zomato workflows with strict pre-order confirmation, cart preview, address checks, and vendor fallback logic.
安装
claude skill add --url https://github.com/openclaw/skills文档
India Food Ordering (Swiggy + Zomato)
What this skill does
Provides one consistent ordering workflow across both vendors:
- Swiggy
- Zomato
Additional vendor targets (when connector/MCP support exists):
- EatSure
- magicpin
- ONDC-compatible food apps
- Blinkit Bistro / Zepto Cafe style quick-food verticals
This skill focuses on safe, human-confirmed ordering operations.
Disclaimer
This skill provides workflow guidance and command orchestration only. It does not guarantee availability, pricing, ETA, or cancellation rights on either platform. Vendor capabilities and policies can change without notice.
Use at your own risk. The skill author/publisher/developer is not liable for direct or indirect loss, wrong orders, payment issues, delays, cancellations, or other damages from use or misuse of this guidance.
Core safety rules (non-negotiable)
- Never place an order without explicit user confirmation
- Always show full cart preview before confirmation
- Always confirm delivery address before placing order
- If COD-only or non-cancellable flows apply, warn user before confirmation
- If uncertain about item, price, or address, ask before proceeding
Standard workflow
Step 1: Collect intent
- cuisine / item
- budget
- location (delivery area)
- delivery speed preference
- preferred vendor (optional)
Step 2: Vendor discovery
- Search both vendors when possible.
- If only one vendor is available, continue with it.
- If both are available, compare and present best options.
Step 3: Compare options
Return side-by-side:
- vendor
- restaurant
- item totals
- delivery fee
- taxes/charges
- ETA
- rating (if available)
Step 4: Build cart
- add selected items
- show running subtotal
- check item availability
Step 5: Mandatory confirmation prompt
Use this exact style:
Ready to place order:
- Vendor: <Swiggy/Zomato>
- Restaurant: <name>
- Items: <list with qty>
- Total payable: <amount>
- Delivery address: <full address label>
- ETA: <time window>
- Notes: <COD only / non-cancellable / special terms if any>
Confirm order? (yes/no)
Step 6: Place order only after "yes"
- Execute order command
- Return order ID/reference immediately
- Share tracking handoff steps
Vendor routing logic
- If user specifies vendor, honor it unless unavailable.
- If unspecified, choose:
- lower final payable
- faster ETA
- higher reliability/rating
- If selected vendor fails, offer fallback on the other vendor and re-confirm.
Error handling
- No restaurants found: broaden radius or cuisine terms.
- Item unavailable: propose equivalent items.
- Cart mismatch: re-fetch cart and re-confirm.
- Auth/session issue: ask user to re-authenticate vendor connector.
- Payment/checkout failure: do not auto-retry order placement; ask user.
Output format
When helping user order, respond in this structure:
- Top options (max 3)
- Recommended vendor and reason
- Cart summary
- Confirmation question
Setup
Read setup.md on first use.
Validation
Run validation-checklist.md before production usage.
References
- Operational notes: vendor-playbook.md
- Prompt examples: examples.md
- Safety and QA checks: validation-checklist.md
- Launch flow: launch-playbook.md
相关 Skills
Claude API
by anthropic
Build, debug, and optimize Claude API / Anthropic SDK apps. Apps built with this skill should include prompt caching. Also handles migrating existing Claude API code between Claude model versions (4.5 → 4.6, 4.6 → 4.7, retired-model replacements). TRIGGER when: code imports `anthropic`/`@anthropic-ai/sdk`; user asks for the Claude API, Anthropic SDK, or Managed Agents; user adds/modifies/tunes a Claude feature (caching, thinking, compaction, tool use, batch, files, citations, memory) or model (Opus/Sonnet/Haiku) in a file; questions about prompt caching / cache hit rate in an Anthropic SDK project. SKIP: file imports `openai`/other-provider SDK, filename like `*-openai.py`/`*-generic.py`, provider-neutral code, general programming/ML.
并行代理
by axelhu
Use when facing 2 or more independent tasks that can be worked on without shared state - dispatches parallel subagents using sessions_spawn for concurrent investigation and execution, adapted for OpenClaw
高光制作器
by bwbernardweston18
>