SandBase
Server Details
Connect MCP clients to 2,000+ AI models without managing provider API keys.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-06-18
- URL
- Repository
- sandbaseai/cli
- GitHub Stars
- 186
- Server Listing
- SandBase
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 7 tools
Most tools map to clearly distinct lifecycle stages: discover, inspect, run, cancel, poll, list, and account check. The main confusion risk is the cluster of similarly named run tools (sandbase_run vs sandbase_run_get vs sandbase_runs), where an agent could pick the listing tool when it wants a single run's status, but the descriptions do clarify the split between execute/poll/list.
All tools share the sandbase_ prefix, but the convention is mixed: some are verb-only (discover, inspect), one is noun-only (account), one is ambiguous (run), and the run family uses resource_action ordering (run_get, run_cancel) alongside a bare plural noun (runs). It remains readable but is not a single predictable pattern.
Seven tools is well-scoped for a model/API marketplace: discovery, inspection, execution, async polling, cancellation, run history, and billing each earn a distinct slot. Nothing feels padded or missing at the count level.
The surface covers the full discover → inspect → run → poll → cancel lifecycle plus run history and balance checks, which is strong coverage. Minor gaps exist around things like vendor/category browsing or managing credentials/keys, but agents can work around these via discover filters.
Available Tools
7 toolssandbase_accountAccountARead-onlyInspect
Check your SandBase account balance. This call is not billed. Use before expensive operations.
Inputs: no named parameters.
sandbase_discoverDiscoverARead-onlyInspect
Find a model, API, or Agent Service for the task. Call this proactively before writing a scraper, before a generic web fetch for structured data, before telling the user something is inaccessible, and whenever you need image/video/music/speech generation, social media data, search results, web scraping, product/company/people data, embeddings, or a specific third-party model. Supports keyword search (q), type filtering (llm/api/multimodal/embedding/service), and vendor filtering. Discovery is not billed; start here.
Inputs: q, type, limit, vendor.
sandbase_inspectInspectARead-onlyInspect
Inspect a model, API, or Agent Service: input schema, available pricing, and an execution template. Service names are svc_ IDs. Always inspect before running.
Inputs: name (required).
sandbase_runRunADestructiveInspect
Execute a model, API, or Agent Service. Sync models return the result directly; async models (image/video/audio jobs) return status running with a run_id — poll sandbase_run_get for the output. Services require arguments.idempotency_key and optionally input; reuse the same key and input on retries. Service acceptance is not completion: poll sandbase_run_get with service_id and run_id. Omit the MCP task parameter for Services.
Inputs: name (required), arguments (required).
sandbase_run_cancelCancel Service RunBDestructiveInspect
Request cancellation of an Agent Service run; inspect its status afterward.
Inputs: run_id (required), service_id (required), idempotency_key (required).
sandbase_run_getGet Run ResultARead-onlyInspect
Get async run status and results. For Agent Services supply both service_id and run_id; succeeded, failed, and cancelled are terminal. Session items are paginated: use cursor for subsequent pages; items_available=false means retry output retrieval. Model runs use completed or failed.
Inputs: run_id (required), limit, cursor, service_id.
sandbase_runsList RunsBRead-onlyInspect
List recent model/API calls, or supply service_id to list runs of an Agent Service.
Inputs: limit, model, offset, service_id.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Every service you connect, one MCP endpoint for all your AI tools. Sign in once.
MCP gateway: one connection for your AI apps, with tool policies, PII redaction and audit logs.
Chat, image, video, and audio models from many vendors behind one remote MCP server and one key.
One MCP endpoint for Claude, GPT & Gemini: 100+ tools + no-code connectors + agent workers.
Related MCP Servers
- AlicenseAqualityBmaintenanceEnables any AI agent to connect to any OpenAI-compatible LLM provider through a unified MCP interface, with automatic failover, caching, cost tracking, and multi-model consensus.13MIT
- AlicenseNot gradedqualityAmaintenanceEnables AI clients to connect to 2,000+ AI models and APIs through the Model Context Protocol, with OAuth, discovery, inspection, execution, pricing checks, and rollback.556,724 npmMIT
- AlicenseBqualityBmaintenanceProvides unified access to 100+ AI models from OpenAI, Anthropic, Google, AWS Bedrock and other providers through a single MCP interface. Enables seamless switching between different AI models using LiteLM's unified API without requiring separate integrations for each provider.312MIT
- AlicenseNot gradedqualityAmaintenanceEnables AI clients to discover, inspect, and run 2,000+ models and APIs through OAuth-backed MCP tools, including free data APIs and paid model calls.556,724 npmMIT
Glama MCP Gateway
Add one secure layer between your agents and this server.