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 Rayyan011/avasfood-mcp --skill avas-food-assistantgit clone --depth 1 https://github.com/Rayyan011/avasfood-mcpWrote 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/rayyan011/avasfood-mcp/avas-food-assistant)<a href="https://agentmods.dev/skills/rayyan011/avasfood-mcp/avas-food-assistant"><img src="https://agentmods.dev/badge/skills/rayyan011/avasfood-mcp/avas-food-assistant/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/rayyan011/avasfood-mcp/avas-food-assistant"><img src="https://agentmods.dev/badge/skills/rayyan011/avasfood-mcp/avas-food-assistant.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00048 | $0.00801 |
| Opus 5 | $0.00024 | $0.00400 |
| Sonnet 5 | $0.00010 | $0.00160 |
| Haiku 4.5 | $0.00005 | $0.00080 |
Grade A, and why
avas-food-assistant 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 11d 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 — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AVAS Food Assistant
Overview
Operate the local avasfood MCP as a conversational food-ordering assistant. Keep browsing casual, but treat receipt upload and order submission as consequential actions requiring separate, explicit confirmation.
When to Use
Use for requests such as:
- “Find rice under MVR 100.”
- “Show food with photos.”
- “Add this and tell me the delivered total.”
- “Upload my transfer slip.”
- “Where is my food?”
Do not use it to guess prices, availability, payment status, or delivery status. Query the MCP.
Workflow
- Discover: call restaurant/menu search tools using the user's saved delivery zone. Present 5–10 useful choices with restaurant, item, current price, and image when available. Completion: every shown item came from the live AVAS response.
- Build locally: add only the items the user selected, then show
view_cart. Completion: the user can see merchant, quantities, and pre-quote subtotal. - Quote: call
quote_cartbefore discussing the final payable amount. Completion: show food subtotal, tax, service charge, delivery/other fees, and grand total separately. - Bank transfer: show the exact AVAS bank details and total. The user performs the transfer outside the agent. Never request banking credentials.
- Receipt: stage the user-selected PNG/JPEG first. Show order amount, merchant, local filename, and image size. Upload only after the user explicitly confirms receipt upload.
- Submit: call
prepare_order, show the immutable preview, and ask for final confirmation. Callsubmit_orderonly with its one-time token after that confirmation. Completion: return AVAS order number or clearly report an ambiguous outcome. - Track: call
track_deliveriesfor live status. Show preparing, rider assigned, out for delivery, ready for pickup, on hold, or delivered; include contacts only when AVAS returns them.
Safety Boundaries
- Searching, menus, addresses, cart inspection, quotes, and tracking are read-only or locally reversible.
- Receipt upload and order submission are separate confirmations.
- Never silently substitute unavailable items, change merchant, change address, or increase quantity.
- Never retry an ambiguous
SaveSalesOrder; inspect active/history orders first. - Do not expose the customer ID, full phone number, full address, session file, tokens, or receipt bytes in chat.
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.
- 11d ago First seen · 66 lines · 48 tokens per session scan A f71c9e8fb8f6
avas-food-assistant is a skill published in the GitHub repository Rayyan011/avasfood-mcp (0 stars, last pushed 1mo ago), licensed MIT. It adds 48 tokens to every session and 801 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
marketplace-purchase-vetting
Use this when the user asks whether a local listing is a scam, "too good to be true," worth looking at, or a good deal. Also use this when he asks you to find options — search/discover candidates, then vet the best ones. The goal is not a generic buying guide; it is a practical risk read with clear next steps.
booth-query-display
Use when the user asks to search/query BOOTH (booth.pm) items, show BOOTH item rankings, or display BOOTH products with covers, CNY prices, and popularity. Covers search, detail fetch, cover images, Japanese-to-Chinese naming, and the fixed display format.
booth-purchase-automation
Buy a real BOOTH item: search, add to cart, confirm, pay, download via browser CDP.
hunt-business-logic
Hunting skill for business logic vulnerabilities. Built from 12 public bug bounty reports. Covers coupon-race-stacking (Instacart, Stripe, Reverb), negative-quantity-in-cart price tampering (Upserve, Eternal/Zomato), decimal/fraction price-field overflow (Shipt), client-side checkout amount trust on PayPal redirect…
foreign-trade-operations
Use this skill when managing a tractor/farm-machinery export business website. Covers: product listing creation, image optimization, SEO for international markets, search engine submission, and site maintenance workflows.
marketing-knowledge-commerce-strategist
A strategy guide for selling educational and expert knowledge through Chinese paid-content platforms. It covers product design, pricing, user operations, personal-brand building, referrals, and business data.