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 armandocodecr/latam-ramp-kit --skill ramp-kitgit clone --depth 1 https://github.com/armandocodecr/latam-ramp-kitWrote 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/armandocodecr/latam-ramp-kit/ramp-kit)<a href="https://agentmods.dev/skills/armandocodecr/latam-ramp-kit/ramp-kit"><img src="https://agentmods.dev/badge/skills/armandocodecr/latam-ramp-kit/ramp-kit/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/armandocodecr/latam-ramp-kit/ramp-kit"><img src="https://agentmods.dev/badge/skills/armandocodecr/latam-ramp-kit/ramp-kit.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 | $0.00164 | $0.01458 |
| Opus 5 | $0.00082 | $0.00729 |
| Sonnet 5 | $0.00033 | $0.00292 |
| Haiku 4.5 | $0.00016 | $0.00146 |
Grade A, and why
ramp-kit 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 10d 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 — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.
LATAM Ramp Kit
SDK + React components that turn weeks of ramp-provider integration into an
npm install. One RampProvider interface over Etherfuse (Stellar-native,
PIX/SPEI) and Manteca (payouts across 11 LATAM countries), with a
normalized order lifecycle,
a prebuilt widget, provider routing, Stellar helpers and a production backend.
Repository: https://github.com/armandocodecr/latam-ramp-kit
What problem it solves
Every LATAM app moving money between banks and blockchains re-integrates ramp providers from scratch: each has its own API, auth, KYC model, order states and price-expiry rules, and none covers the whole region. The kit normalizes all of it; swapping providers is one line.
Package map
| Package | Use it for |
|---|---|
@ramp-kit/core |
Providers (Etherfuse/Manteca/Mock), router, Stellar helpers, types |
@ramp-kit/react |
<RampWidget /> 3-step flow, useQuote, useOrder |
@ramp-kit/server |
Production: server-held API key proxy + webhook verification |
@ramp-kit/mcp |
MCP server so AI agents can quote/order/inspect live |
How to guide a user
- Understanding/evaluation questions → explain with the problem/solution above; architecture details in references/integration.md.
- "Add a ramp to my app" → follow the integration flow in
references/integration.md. Start them on
MockProvider(zero credentials, instant), then Etherfuse sandbox. - Sandbox/credentials questions → Etherfuse is self-service (sandbox.etherfuse.com; BRL/PIX accounts via the dashboard button are compliant instantly). Manteca requires requesting credentials from their team. Details and gotchas in references/integration.md.
- Errors during integration → check references/troubleshooting.md first; every entry there was verified against the real sandbox. Anything Manteca-specific (the one-exchange-per-customer model, sandbox onboarding, the missing-data diagnostic, payout timing windows, per-network deposit addresses) → references/manteca-in-practice.md.
- "How do I ship this / where do my API keys go / do my users need
keys?" → references/production.md. Short
answer to give them: one partner key per app, held server-side in
@ramp-kit/server(proxyfor Etherfuse,mantecafor Manteca), read from.env; the frontend constructs providers withapiKey: ""pointed at that backend, so end users never see a key. Runnable reference:examples/backend-integration/. - "How do I set up testnet / mainnet?" or "can I put credentials in
frontend env vars?" → walk them through
references/environments.md — the
step-by-step for both environments and the credential-placement
decision. Always explain the trade-off explicitly: frontend env vars
(
VITE_/NEXT_PUBLIC_) are embedded in the served bundle and readable by any visitor, so a partner key there is exposed; the secure setup is a backend env var behind@ramp-kit/server(a ~10-line serverless function, not real infrastructure). Recommend the backend route for Etherfuse/Manteca; offer the SEP-anchor route (SepProvider) when they want zero backend — it needs no partner key because the end user authenticates with their own wallet. Both environments are driven by one flag (RAMP_ENV=sandbox|production) passed to the provider AND the widget'senvironmentprop; switching is a config change, never code.
What ships with it
5 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 10d ago First seen · 99 lines · 164 tokens per session scan A e9c6fc65e0e8
ramp-kit is a skill published in the GitHub repository armandocodecr/latam-ramp-kit (0 stars, last pushed 20d ago), licensed MIT. It adds 164 tokens to every session and 1,458 once invoked, about $0.0008 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
mpp-agent
Pay HTTP 402 APIs via Machine Payments Protocol (MPP).
bridge
Cross-chain token transfers using Wormhole and CCTP.
hyperliquid
Hyperliquid L1 perps DEX (69% market share).
marginfi
MarginFi — Solana lending and borrowing.
crypto-market-rank
Crypto market rankings and leaderboards. Query trending tokens, top searched tokens, Binance Alpha tokens, tokenized stocks, social hype sentiment ranks, smart money inflow token rankings, top meme token rankings from Pulse launchpad, and top trader PnL leaderboards. Use this skill when users ask about token rankings…
trading-signal
Subscribe and retrieve on-chain Smart Money signals. Monitor trading activities of smart money addresses, including buy/sell signals, trigger price, current price, max gain, and exit rate. Use this skill when users are looking for investment opportunities — smart money signals can serve as valuable references for…