Rhumb
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| find_toolsA | Semantic search for agent tools, ranked by AN Score |
| get_scoreB | Get detailed AN Score breakdown for a service |
| get_alternativesC | Find alternative services ranked by AN Score |
| get_failure_modesC | Get known failure patterns for a service |
| discover_capabilitiesA | Discover what capabilities are available — search by domain or text. Returns capabilities with provider counts and top provider info. |
| resolve_capabilityA | Resolve a capability to ranked providers with health-aware recommendations, costs, and fallback chains. The core agent decision: 'I need email.send — what should I use?' |
| execute_capabilityA | Execute a capability through Rhumb. Three credential modes: (1) byo — bring your own token, requires method+path; (2) rhumb_managed — zero-config, Rhumb provides credentials, method/path optional; (3) agent_vault — pass your own token via agent_token param (get it from credential_ceremony first). Use resolve_capability to see providers and check_credentials to see what modes are available. |
| estimate_capabilityA | Get cost estimate for executing a capability without actually executing it. Use before expensive operations or when building cost-aware workflows. |
| credential_ceremonyA | Get step-by-step instructions for obtaining API credentials for a service. Without a service param, lists all available ceremonies. With a service param, returns detailed steps, token format info, and documentation links. Use this before agent_vault execute mode. |
| check_credentialsA | Check your credential status across all three modes. Shows which capabilities are available via Rhumb-managed (zero-config), which services have ceremony guides for self-provisioning, and BYO status. Start here to understand what you can execute. |
| budgetA | Check or set your execution budget. Without action param, returns current status. Use action='set' with budget_usd to create/update. Budget is enforced pre-execution — you get 402 (not a surprise bill) when over limit. |
| spendA | Get your spend breakdown by capability and provider. Shows total spend, execution count, and cost averages. Defaults to current month. |
| routingA | Get or set your provider routing strategy. Strategies: cheapest (lowest cost above quality floor), fastest (healthiest circuits), highest_quality (highest AN score), balanced (weighted mix). Quality floor filters out low-quality providers. |
| check_balanceB | Check the current credit balance for your organization on Rhumb |
| get_payment_urlA | Get a Stripe checkout URL to top up Rhumb credits |
| get_ledgerC | Get recent billing ledger entries for your organization |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 16 tools
Most tools have distinct purposes, but some overlap exists: 'discover_capabilities' and 'find_tools' both involve searching for capabilities/tools, which could cause confusion. However, their descriptions clarify that 'discover_capabilities' is domain-focused while 'find_tools' is semantic search for agent tools, helping to mitigate ambiguity.
Tool names follow a consistent verb_noun pattern throughout, such as 'check_balance', 'discover_capabilities', and 'execute_capability'. All names use snake_case and clear, descriptive verbs, making the set predictable and easy to understand.
With 16 tools, the count is slightly high but reasonable for a comprehensive platform like Rhumb that manages capabilities, credentials, billing, and routing. It covers multiple aspects without feeling overly bloated, though it borders on the upper limit of a well-scoped set.
The tool set provides complete coverage for Rhumb's domain, including capability discovery, execution, credential management, cost estimation, billing, and routing strategies. There are no obvious gaps; agents can perform end-to-end workflows from setup to execution and monitoring without dead ends.