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 davidmosiah/rappi-mcp --skill skillgit clone --depth 1 https://github.com/davidmosiah/rappi-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/davidmosiah/rappi-mcp/skill)<a href="https://agentmods.dev/skills/davidmosiah/rappi-mcp/skill"><img src="https://agentmods.dev/badge/skills/davidmosiah/rappi-mcp/skill/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/davidmosiah/rappi-mcp/skill"><img src="https://agentmods.dev/badge/skills/davidmosiah/rappi-mcp/skill.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 | $0.00071 | $0.00573 |
| Opus 5.5 | $0.00028 | $0.00229 |
| Sonnet 5.5 | $0.00014 | $0.00115 |
| Haiku 4.5 | $0.00007 | $0.00057 |
Grade A, and why
rappi 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 1mo 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.
What it actually says
Rappi — skill or MCP
Unofficial. Not the Rappi Partners restaurant API.
Same binary either way. Mutation gates live in the server, not in this file.
Choose a surface
MCP — tools appear natively:
{ "mcpServers": { "rappi": { "command": "npx", "args": ["-y", "rappi-mcp-unofficial"] } } }
Do not put mutation flags in that snippet.
Skill / CLI — no MCP client required:
npx -y rappi-mcp-unofficial doctor --json
npx -y rappi-mcp-unofficial call rappi_capabilities --json '{}'
npx -y rappi-mcp-unofficial call rappi_search_stores --json '{"query":"leite"}'
If MCP tools named rappi_* are already available, use them. Do not also shell out.
Setup (once)
npx -y rappi-mcp-unofficial setup
npx -y rappi-mcp-unofficial auth --from-header "Bearer …"
Token is a captured Authorization from services.rappi.com.br, stored at ~/.rappi-mcp/tokens.json (0600). Guest search works without a token; checkout stays blocked.
Loop
rappi_connection_status(ordoctor --json). Expectunofficialandnever_pays_by_default.- Search / home / cart / orders as asked. Street, phone, GPS, receipt images stay redacted.
- “O que tem no carrinho?” →
rappi_get_cart. “Pedir o de sempre” →rappi_list_ordersthenrappi_checkout_preview, stop.rappi_reorderonly if the user named that order and both gates are already on. - Stop before pay. Do not call
rappi_place_order(or other gated pay/cart writes) unless the user clearly asked to place this order. If the tool returnsUSER_ACTION_REQUIRED, report that and stop. Do not invent env flags.
Never
- Enable mutations from this skill
- Paste tokens into git, chat logs, or the prompt
- Treat guest as able to pay
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.
- 1mo ago Changed · +1 lines dcb5ea04d66b
- 1mo ago First seen · 57 lines · 71 tokens per session scan A 595766611d86
rappi is a skill published in the GitHub repository davidmosiah/rappi-mcp (0 stars, last pushed 2d ago), licensed MIT. It adds 71 tokens to every session and 573 once invoked, about $0.0003 per session on Opus 5.5. 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
seller-research
Use when researching a merchant, storefront, marketplace seller, or merchant of record for a buying decision, especially when identity, refund terms, fulfillment, counterfeit risk, domain history, or independent buyer outcomes are uncertain.
ze
Unofficial Zé Delivery (Brazil) for catalog, product search and order tracking. Use when the user wants Zé without opening the app. Prefer MCP tools if connected; otherwise the package CLI. Never pays unless both gates are already on and the user said to place this order.
ifood
Unofficial iFood (Brazil) for personal restaurants, cart, addresses and orders. Use when the user wants iFood without opening the app. Prefer MCP tools if connected; otherwise the package CLI. Never pays unless both gates are already on and the user said to checkout.
Product Photoshoot Workflow
A workflow for turning one product photo into a set of four e-commerce images from different visual directions.
escalation
Escalate a disputed refund to the billing lead.
sfcc-sfra-scss
Best practices for styling and theming SFRA storefronts using SCSS. Use when asked to create style overrides, theming, responsive layouts, or CSS customizations in SFCC.