mp-integrate
Wizard that scaffolds a complete Mercado Pago integration for any product. Asks the developer the minimum questions needed (country, product, variant, SDK, mode), queries the MCP server for live docs, and produces a ready-to-paste code bundle. Use whenever the developer wants to add or migrate a Mercado Pago payment flow.
What this skill does
# mp-integrate
This skill is the single entry point for building a Mercado Pago integration. It collects the minimum context from the developer, queries the official Mercado Pago MCP server (`plugin:mercadopago:mcp`) for current documentation, and assembles a ready-to-paste bundle (server snippet + client snippet + env vars + test instructions + gotchas).
**The MCP server is the source of truth.** This skill orchestrates queries to it; it does not duplicate documentation. If `mcp__plugin_mercadopago_mcp__search_documentation` is not available, stop and instruct the user to run `/mp-connect`.
---
## Step 0 — Verify MCP is actually authenticated
`ListMcpResourcesTool` is **not** a reliable check — this MCP returns "No resources found" whether authenticated or not. The bootstrap tools `authenticate` / `complete_authentication` are **always** present and prove nothing.
The reliable check: is `mcp__plugin_mercadopago_mcp__application_list` callable from the current tool list AND does it return at least one application?
- If the tool is not in your capabilities (only `authenticate` / `complete_authentication` are visible) → call `mcp__plugin_mercadopago_mcp__authenticate` immediately. Show the URL and say:
> Open this URL to connect Mercado Pago: **{authorization_url}**
> When you see **"Authentication Successful"** in the browser, come back and say anything.
When the user responds, **call `application_list` directly** — the local MCP server already processed the callback when the browser hit the redirect URL. Do NOT call `complete_authentication` first (it will hang trying to open a socket that's already closed). Only fall back to `complete_authentication` if `application_list` still fails AND the browser showed an error, not "Authentication Successful". Never ask the user to paste the callback URL — it contains a sensitive OAuth code.
- **`application_list` callable and returns an app** → authenticated. Save the response and continue.
- **Only `authenticate`/`complete_authentication` visible** → loaded but not authenticated. Call `authenticate`, show the URL as a clickable link, wait for the user to return, then call `application_list` directly. **Do NOT call `complete_authentication`** — it hangs because the local server already consumed the callback.
- **Neither tool visible** → plugin is disabled or not loaded. Tell the user: *"The Mercado Pago plugin isn't loaded. Run `/mcp` in the terminal, find `plugin:mercadopago:mcp`, and enable it. Then run `/mp-integrate` again."* Do NOT suggest `/mp-connect` — it also requires the plugin to be loaded.
---
## ⚠️ HARD LOCKS — read before doing anything else
These rules override any "what makes sense" judgement during the wizard. Past wizard runs have violated them; do not repeat the mistake.
### LOCK 1 — SDK is never a wizard question
The SDK / language is **NEVER** asked via `AskUserQuestion`. Period.
- Resolve it silently in Step 1.a by globbing the repo for a manifest (`package.json`, `pyproject.toml`, etc.).
- If a single manifest is found → record the SDK and **skip the question entirely**.
- If multiple manifests exist (real polyglot monorepo) → still don't ask. Pick the one that matches the directory the developer is currently editing, or default to `node`. Mention the choice in a single line of chat (`✓ SDK: node — from package.json`) and **continue without asking**.
- If no manifest exists at all → still don't ask. Default to `node`, mention it (`✓ SDK: node — defaulted (no manifest detected; we'll create package.json during scaffolding)`), and continue.
If you find yourself about to call `AskUserQuestion` with `header="SDK"` or `header="Stack"` or `header="Language"`, **stop immediately**. The SDK is never a picker. The Tabs row at the top of the wizard must NOT include "SDK" as one of the tabs.
### LOCK 2 — Product → Mode availability table (NON-NEGOTIABLE)
| Product | The ONLY valid `mode` values | Picker behavior |
|---------|------------------------------|-----------------|
| `checkout-pro` | `preferences` (the Orders API does **not** exist for Checkout Pro) | **Skip the mode question entirely.** Do not call `AskUserQuestion` with `header="Mode"`. Do not show "Orders API" as an option. Use `mode=preferences` silently. |
| `checkout-api` | `orders`, `payments` | Ask only if `Grep` did not disambiguate. Default `orders`. |
| `bricks` | `orders` | Skip the mode question (single value). |
| `qr` | `orders`, `legacy` | Ask only if ambiguous. |
| `point` | `orders`, `legacy` | Ask only if ambiguous. |
| `marketplace` | `orders`, `legacy` | Ask only if ambiguous. |
| `wallet-connect` | `orders` | Skip the mode question (single value). |
| `subscriptions` | n/a (uses its own `preapproval` API) | Skip the mode question. |
| `money-out` | n/a (uses its own `disbursements` API) | Skip the mode question. |
| `smartapps` | n/a | Skip the mode question. |
**If `product=checkout-pro` and you are about to render a Mode picker that includes `Orders API`, abort.** The Orders API is not available for Checkout Pro today. Period. Do not "future-proof" by offering it. Do not add a "Recommended" tag to it. Do not include it in any "Other" fallback.
### LOCK 3 — Always use `init_point`, never `sandbox_init_point`
Mercado Pago removed the sandbox environment. There is no staging URL. Every integration — including test-user flows — runs against the production API.
**Never generate code that references `sandbox_init_point`.** Always use `init_point` from the preference response. The difference between a test run and a production run is only which credentials are loaded (`APP_USR-` from a test user vs. a real account) — not the URL.
If you find `sandbox_init_point` in existing code, flag it as a bug: the redirect will fail silently or land on an error page.
**Also applies to test users:** test users created via `create_test_user` operate against the production API using `APP_USR-` credentials. There is no separate test base URL, no `TEST-` prefix, and no toggle. Code using `sandbox_init_point` with test-user credentials will not work.
### LOCK 4 — Tabs row must reflect only the questions that will actually be asked
The wizard's Tabs row at the top (the `□ Country □ Product □ Mode ✓ Submit` line) must include **only** the dimensions that are actually still unresolved AND non-skipped per LOCK 1 and LOCK 2. Concretely:
- Never include "SDK" in the tabs — see LOCK 1.
- Never include "Mode" in the tabs when `product` is `checkout-pro` / `bricks` / `wallet-connect` / `subscriptions` / `money-out` / `smartapps`.
- Never include "Environment" in the tabs.
---
## Step 1 — Parse `$ARGUMENTS` and ask for missing context
`$ARGUMENTS` may include any combination of these flags. Anything missing must be asked via `AskUserQuestion` in batches of ≤4.
| Flag | Values |
|------|--------|
| `country=` | `AR` / `BR` / `MX` / `CL` / `CO` / `PE` / `UY` |
| `product=` | `checkout-pro` / `checkout-api` / `bricks` / `qr` / `point` / `subscriptions` / `marketplace` / `wallet-connect` / `money-out` / `smartapps` |
| `mode=` | depends on product — see Product Matrix below |
| `sdk=` | `node` / `python` / `java` / `php` / `ruby` / `dotnet` / `go` (or `none` for raw REST) |
| `client=` | `vanilla-js` / `react` / `ios` / `android` / `flutter` / `react-native` (only for products with a client component) |
| `lang=` | `es` / `en` / `pt` (docs language) |
| `recurrent=` | `yes` / `no` (Checkout API, Bricks) |
| `3ds=` | `yes` / `no` (Checkout API, Bricks) |
| `marketplace=` | `yes` / `no` (split payments) |
| `brick=` | `payment` / `card-payment` / `wallet` / `status-screen` (only when `product=bricks`) |
| `qr-mode=` | `static` / `dynamic` / `attended` (only when `product=qr`) |
### Step 1.a — Auto-resolve before asking (MANDATORY — exhaust this step first)
**You MUST run this step before any `AskUserQuestion` call.** Every dimension that resolves here is removed from the wizard. The developer should only be askeRelated in Backend & APIs
jfrog
IncludedInteract with the JFrog Platform via the JFrog CLI and REST/GraphQL APIs. Use this skill when the user wants to manage Artifactory repositories, upload or download artifacts, manage builds, configure permissions, manage users and groups, work with access tokens, configure JFrog CLI servers, search artifacts, manage properties, set up replication, manage JFrog Projects, run security audits or scans, look up CVE details, query exposures scan results from JFrog Advanced Security, manage release bundles and lifecycle operations, aggregate or export platform data, or perform any JFrog Platform administration task. Also use when the user mentions jf, jfrog, artifactory, xray, distribution, evidence, apptrust, onemodel, graphql, workers, mission control, curation, advanced security, exposures, or any JFrog product name.
cupynumeric-migration-readiness
IncludedPre-migration readiness assessor for porting NumPy to cuPyNumeric. Use BEFORE substantial porting work begins when the user asks whether code will scale on GPU, whether they should migrate to cuPyNumeric, which NumPy patterns transfer cleanly, what must be refactored before porting, or mentions pre-port assessment, scaling analysis, or refactor planning. Inspect the user's source code, look up NumPy usage, cross-reference the cuPyNumeric API support manifest, and distinguish distributed-scaling-friendly patterns from blockers such as unsupported APIs, scalar synchronization, host round-trips, Python/object-heavy control flow, shape/data-dependent branching, and in-place mutation hazards. Produce a verdict of READY, LIGHT REFACTOR, SIGNIFICANT REFACTOR, or NOT RECOMMENDED, with concrete refactor pointers.
alibabacloud-data-agent-skill
IncludedInvoke Alibaba Cloud Apsara Data Agent for Analytics via CLI to perform natural language-driven data analysis on enterprise databases. Data Agent for Analytics is an intelligent data analysis agent developed by Alibaba Cloud Database team for enterprise users. It automatically completes requirement analysis, data understanding, analysis insights, and report generation based on natural language descriptions. This tool supports: discovering data resources (instances/databases/tables) managed in DMS, initiating query or deep analysis sessions, real-time progress tracking, and retrieving analysis conclusions and generated reports. Use this Skill when users need to query databases, analyze data trends, generate data reports, ask questions in natural language, or mention "Data Agent", "data analysis", "database query", "SQL analysis", "data insights".
token-optimizer
IncludedReduce OpenClaw token usage and API costs through smart model routing, heartbeat optimization, budget tracking, and native 2026.2.15 features (session pruning, bootstrap size limits, cache TTL alignment). Use when token costs are high, API rate limits are being hit, or hosting multiple agents at scale. The 4 executable scripts (context_optimizer, model_router, heartbeat_optimizer, token_tracker) are local-only — no network requests, no subprocess calls, no system modifications. Reference files (PROVIDERS.md, config-patches.json) document optional multi-provider strategies that require external API keys and network access if you choose to use them. See SECURITY.md for full breakdown.
resend-cli
IncludedUse this skill when the task is specifically about operating Resend from an AI agent, terminal session, or CI job via the official resend CLI: installing/authenticating the CLI, sending/listing/updating/cancelling emails, batch sends, domains and DNS, webhooks and local listeners, inbound receiving, contacts, topics, segments, broadcasts, templates, API keys, profiles, or debugging Resend CLI/API failures. Trigger on mentions of Resend CLI, `resend`, `resend doctor`, `resend emails send`, `resend domains`, `resend webhooks listen`, `resend emails receiving`, or agent-friendly terminal automation.
alibabacloud-odps-maxframe-coding
IncludedUse this skill for MaxFrame SDK development and documentation navigation on Alibaba Cloud MaxCompute (ODPS). Helps answer MaxFrame API, concept, official example, and supported pandas API questions; create data processing programs; read/write MaxCompute tables; debug jobs (remote or local); and build custom DPE runtime images. Trigger when users mention MaxFrame, MaxCompute with MaxFrame, ODPS table processing, DPE runtime, MaxFrame docs/examples, DataFrame/Tensor operations, or GPU runtime setup. Works for both English and Chinese queries about Alibaba Cloud data processing with MaxFrame.