Skip to content
Back to skills

Ramp Install Auth

ASecurity

Select and configure the correct Ramp OAuth grant, scopes, environments, and secret lifecycle. Use when bootstrapping an internal server integration or a third-party multi-tenant application. Trigger with "set up Ramp OAuth" or "Ramp client credentials".

  • 2,719 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added September 21, 2026
ai-agentsrusttestingapisecuritydocumentation

Works with

  • cli
  • api

Security analysis

A100/100

Pro scans all 2 files and shows the line behind each finding

Scanned September 21, 2026

npx -y skills add jeremylongshore/claude-code-plugins-plus --skill ramp-install-auth --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Ramp Install Auth?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Ramp Install Auth
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/jeremylongshore-ramp-install-auth-e5bdb464/badge)](https://www.skillsdirectory.com/skills/jeremylongshore-ramp-install-auth-e5bdb464)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: ramp-install-auth
description: >-
  Select and configure the correct Ramp OAuth grant, scopes, environments, and secret lifecycle. Use when bootstrapping an internal server integration or a third-party multi-tenant application. Trigger with "set up Ramp OAuth" or "Ramp client credentials".
allowed-tools: Read,Glob,Grep,Write,Edit
version: 2.0.0
argument-hint: "[integration-or-environment]"
model: inherit
effort: high
author: Jeremy Longshore <jeremy@intentsolutions.io>
license: MIT
compatibility: Requires current Ramp Developer API documentation and approved access for any live financial, card, identity, accounting, application, or configuration change
tags: [saas, ramp, oauth, authentication, installation]
---
# Ramp OAuth Application and Authentication Setup

## Overview

Choose the grant from the trust model. Use Client Credentials for an internal server-to-server integration; use Authorization Code with state and refresh handling for a third-party or multi-tenant application.

## Prerequisites

- Identify the Ramp application, environment, business entities, affected data and workflows, accountable owner, and rollback boundary.
- Read `references/official-docs.md` and re-check endpoint schemas, scopes, limits, and support status before a live operation.
- Use synthetic fixtures or Ramp sandbox until production access and business effects are explicitly approved.
- Prepare approved secret storage and a sanitized evidence location.

## Current Contract

- Ramp's token endpoint is `/developer/v1/token`, typically using HTTP Basic authentication for the client.
- Client Credentials tokens last 10 days; Authorization Code and refresh-token access tokens last one hour according to the current authorization guide.
- Authorization Code is required for third-party integrations acting for other businesses; its authorization code is short-lived and redirect URI matching is exact.
- Tokens are opaque, environment-bound, and scope-bound; changing configured scopes requires a new token.

## Instructions

1. Identify internal versus third-party architecture, target environments, tenants, redirect origins, required operations, data classes, and accountable application owner.

2. Register separate sandbox and production applications; enable only the selected grant and minimum endpoint scopes.

3. Store client secrets and refresh tokens in an approved secret manager, record rotation/revocation ownership, and prohibit browser or repository exposure.

4. For Authorization Code, generate and verify cryptographically random state and exact HTTPS redirect URIs; for Client Credentials, keep token exchange server-side.

5. Acquire a sandbox token, verify one minimum-scope read plus one denied operation, then document the independently approved production promotion.

## Tool Discipline

- Use **Glob** to locate candidate code, manifests, fixtures, and evidence without widening scope.
- Use **Grep** to find relevant endpoints, fields, permissions, identifiers, errors, and stale assumptions.
- Use **Read** to inspect the smallest required local files and authoritative evidence.
- Use **Write** only for a new approved local draft, test, configuration, or evidence artifact.
- Use **Edit** only for a bounded approved change with a known rollback.
- Local file tools do not authorize a Ramp operation or replace owner approval.

## Approval Boundaries

The business owner approves the application; security approves grant, redirects, scopes, and secret lifecycle; each tenant's authorized user grants third-party access.

## Output

An application inventory, architecture/grant decision, scope matrix, redirect and environment map, secret references, token lifecycle, positive/negative proof, and revocation procedure.

## Error Handling

| Condition | Response |
|---|---|
| `invalid_client` | Verify environment-specific client credentials and Basic authentication without exposing values. |
| `invalid_scope` | Align configured and requested scopes, obtain a new token, and repeat negative testing. |
| `Business not authorized` | Have an appropriately privileged business user authorize the app; do not bypass user authority. |

## Examples

### Example 1

Configure an internal accounting worker with Client Credentials and only transaction/accounting scopes in sandbox.

### Example 2

Configure a multi-tenant partner app with Authorization Code, exact redirects, verified state, encrypted refresh tokens, and tenant revocation.

## Validation

- Grant type matches the internal or third-party architecture.
- Sandbox and production use separate apps, hosts, credentials, and evidence.
- Every scope maps to a documented endpoint and owner.
- Secrets are absent from source and logs; rotation, revocation, expiry, and negative tests work.

## Resources

- [Official documentation and contract notes](references/official-docs.md)
- Re-check the dated contract and current OpenAPI schema before any live request.
- Treat unresolved vendor behavior, authority, or financial state as a stop condition.

Files in this skill

  • SKILL.md4.9 KB
  • references/official-docs.md871 B

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.