Welcome to swarmvia
Introduction
swarmvia is the production context layer for agents. It gives agents and teammates safe, team-managed access to live MCP servers, PostgreSQL or MySQL databases, and APIs. They work from real tools, tables, and columns for more accurate results, without putting passwords or tokens on every laptop. Add each one as a source. swarmvia holds its credential, checks who may use it, logs every call, and admits in-flight requests up to your concurrency limit.
Agents connect at https://swarmvia.com/mcp.
There is no token to paste. Scripts call the source's endpoint on
*.swarmvia.cloud or take a concurrency lock. HTTPS APIs
use the same source model when a vendor or your own service has a hard shared
limit.
Proxy when you want swarmvia to call upstream. Use locks when you front a public endpoint or keep the origin client in your process. For SaaS origins with account-level concurrency limits, see the knowledge base.
Alongside your gateway
Keep your API gateway out front for auth, routing, and WAF. Put swarmvia closest to the databases, MCP servers, and APIs your agents and teammates need, so swarmvia decides who sees each one, holds the credential, logs every call, and shares one limit across all callers.
How it fits together
- Create a source and point it at one or more allowed origins.
- Choose managed custody (swarmvia holds and signs upstream credentials) or pass-through (your client sends upstream auth; swarmvia relays it).
- Mint an access key and call
https://‹source›.swarmvia.cloud/...with that key. - Tune concurrency, queue depth, RPM, retries, and optional response cache on the source's Concurrency settings.
- When several sources share one origin concurrency limit, put them on a Capacity budget. Fixed caps each source at its allocation. Auto keeps reserved pins local and moves float toward load.
- Use the source's Metrics tab for admit/queue/reject proof. Billing remains the place for request allowance.