Stats
Actions
Tags
From ramp-pack
Lists Ramp virtual cards with limits/status, fetches recent transactions by merchant/category, and retrieves users by role via API. For corporate card/expense management.
How this skill is triggered — by the user, by Claude, or both
Slash command
/ramp-pack:ramp-hello-worldThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
List cards, get transactions, and check user details using the Ramp API.
List cards, get transactions, and check user details using the Ramp API.
ramp-install-auth with valid access tokenresp = requests.get(f"{BASE}/cards", headers=headers, params={"page_size": 10})
for card in resp.json()["data"]:
print(f"Card: {card['display_name']} — Limit: ${card['spending_restrictions']['amount']/100:.2f}")
print(f" Status: {card['state']}, Last4: {card['last_four']}")
resp = requests.get(f"{BASE}/transactions", headers=headers, params={
"start_date": "2026-01-01",
"page_size": 10,
})
for tx in resp.json()["data"]:
print(f"${tx['amount']/100:.2f} at {tx['merchant_name']} — {tx['sk_category_name']}")
resp = requests.get(f"{BASE}/users", headers=headers, params={"page_size": 10})
for user in resp.json()["data"]:
print(f" {user['first_name']} {user['last_name']} — {user['role']}")
| Error | Cause | Solution |
|---|---|---|
| Empty results | No data in sandbox | Create test cards first |
403 Forbidden | Insufficient permissions | Check API app permissions |
400 Bad date format | Wrong date format | Use ISO 8601: YYYY-MM-DD |
Issue virtual cards: ramp-core-workflow-a
npx claudepluginhub jeremylongshore/claude-code-plugins-plus-skills --plugin ramp-packAutomates corporate finance operations via Ramp platform: retrieve transactions, manage reimbursements, search expenses, view cards, and list users for expense management and accounting workflows.
Implements Python patterns for Ramp API integration to fetch corporate cards, handle expenses, and sync accounting data using OAuth2 authentication.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.