What can you do with this server?

Rhumb is an agent-native tool intelligence server for discovering, evaluating, and executing external services with trust scores, failure analysis, cost-aware routing, and credential management. Discovery & Evaluation * find_tools — Semantic search for agent tools ranked by AN Score (0–10 quality rating) * get_score — Detailed AN Score breakdown (execution quality, access readiness, autonomy support) * get_alternatives — Ranked alternatives for any service * get_failure_modes — Known failure patterns before integration * discover_capabilities — Browse 414 capabilities across 23 providers by domain or text Execution & Routing * resolve_capability — Resolve a capability (e.g. email.send) to ranked, health-aware providers with costs and fallback chains * execute_capability — Execute a capability via BYO, Rhumb-managed, or Agent Vault credentials * estimate_capability — Cost estimate before committing to execution * routing — Configure provider selection strategy: cheapest, fastest, highest_quality, or balanced Credential Management * credential_ceremony — Step-by-step instructions for obtaining API credentials * check_credentials — Check credential status across all three modes Budget & Billing * budget — Set and enforce spending limits (hard/soft controls) * spend — Spending breakdown by capability and provider * check_balance — Current prepaid credit balance * get_payment_url — Stripe checkout URL to top up credits * get_ledger — Recent billing ledger entries (debits, credits, auto-reloads)

Which integrations are available for this server?

Enables agents to discover, score, and execute Stripe capabilities with visibility into failure modes, execution quality, and cost-aware routing.

Rhumb

npm version License: MIT MCP Registry

Agent-native tool intelligence. Discover, evaluate, and execute external tools — with trust scores, failure modes, cost-aware routing, and managed credentials.

🌐 rhumb.dev · ⚡ Quickstart · 💵 Pricing · 📊 Leaderboard · 📖 Methodology · 🔑 Trust

For agents: See llms.txt for machine-readable documentation and agent-capabilities.json for structured capability metadata.

Start in 30 seconds

MCP (recommended)

npx rhumb-mcp@latest

Zero config. Discovery tools work immediately — no signup, no API key.

For execution, pass your Rhumb API key:

RHUMB_API_KEY=your_key npx rhumb-mcp@latest

Get an API key →

API (read-only, no auth)

curl "https://api.rhumb.dev/v1/services/stripe/score"

All read endpoints are public.

What Rhumb does

Agents need external tools. Choosing the right one is hard — not because of feature lists, but because of:

auth and signup friction
provisioning reality vs. marketing claims
schema instability
failure recovery when no human is watching
hidden costs and rate limits

Rhumb makes those constraints visible before you commit.

Rhumb Index — Discover & Evaluate

695 scored services across 50+ domains. Each gets an AN Score (0–10) measuring execution quality, access readiness, and agent autonomy support.

find_services — search by capability, domain, or name
an_score — full score breakdown with dimension-level detail
get_alternatives — ranked alternatives for any service
get_failure_modes — known failure patterns before you integrate

Rhumb Resolve — Execute

414 capabilities across 23 providers. Cost-aware routing picks the best provider for each call.

execute_capability — call a capability through Resolve with managed auth
resolve_capability — see ranked providers before executing
estimate_capability — get cost estimate before committing
Budget enforcement, credential management, and execution telemetry included

Three credential modes

Mode	How it works
BYO	Bring your own API key — Rhumb routes, you authenticate
Rhumb-managed	Rhumb holds the credential — zero setup for the agent
Agent Vault	Your key, encrypted and stored — Rhumb injects at call time

Payment paths

API key — sign up, get a key, prepaid credits
x402 / USDC — no signup, pay per call on-chain

MCP tools

rhumb-mcp exposes 17 tools:

Discovery

find_services — search services
an_score — score breakdown
get_alternatives — ranked alternatives
get_failure_modes — failure patterns
discover_capabilities — browse capability registry
usage_telemetry — your usage analytics

Execution

resolve_capability — ranked providers for a capability
execute_capability — execute through Resolve
estimate_capability — cost estimate before execution
credential_ceremony — set up credentials
check_credentials — verify credential status
routing — configure routing strategy

Billing

budget — set spend limits
spend — check current spend
check_balance — prepaid balance
get_payment_url — top-up link
get_ledger — transaction history

API

Base URL: https://api.rhumb.dev/v1

Endpoint	Auth	Purpose
`GET /services/{slug}/score`	No	Score breakdown
`GET /services/{slug}`	No	Service profile + metadata
`GET /services/{slug}/failures`	No	Known failure modes
`GET /search?q=...`	No	Search services
`GET /leaderboard/{category}`	No	Category rankings
`GET /capabilities`	No	Capability registry
`GET /capabilities/{id}/resolve`	No	Ranked providers
`POST /capabilities/{id}/execute`	Yes	Execute a capability
`GET /capabilities/{id}/execute/estimate`	Yes	Cost estimate
`GET /telemetry/provider-health`	No	Provider health status
`GET /telemetry/usage`	Yes	Your usage analytics
`GET /pricing`	No	Machine-readable pricing

Examples

See examples/ for runnable scripts:

Example	What it shows	Auth needed?
discover-and-evaluate.py	Search → Score → Failure modes	No
resolve-and-execute.py	Resolve → Estimate → Execute	Yes
budget-aware-routing.py	Budget + cost-optimal routing	Yes
dogfood-telemetry-loop.py	Repeatable Resolve → telemetry verification loop	Yes
mcp-quickstart.md	MCP setup for Claude, Cursor, etc.	Optional

# Try discovery right now (no auth needed)
pip install httpx && python examples/discover-and-evaluate.py

Docs

Agent Accessibility Guidelines — making web interfaces usable by AI agents
AN Score Methodology — scoring dimensions, weights, and rubrics
Architecture — scoring engine design
API Reference — endpoint details
Security Policy — vulnerability reporting and security architecture
Runbooks — operational procedures
Dogfood Loop — repeatable Resolve → telemetry validation harness

Repo structure

rhumb/
├── packages/
│   ├── api/         # Python API (Railway)
│   ├── astro-web/   # Public website (Vercel)
│   ├── mcp/         # MCP server (npm)
│   ├── cli/         # CLI tooling
│   └── shared/      # Shared types/constants
├── examples/        # Runnable examples
├── docs/            # Public documentation
├── scripts/         # Scoring + verification
├── artifacts/       # Score datasets
├── llms.txt         # Machine-readable docs for agents
└── agent-capabilities.json  # Structured capability manifest

Development

# API
cd packages/api && pip install -r requirements.txt && uvicorn app:app --reload

# MCP
cd packages/mcp && npm ci && npm run dev

# Web
cd packages/astro-web && npm ci && npm run dev

Node 24+ recommended (.nvmrc included).