> ## Documentation Index
> Fetch the complete documentation index at: https://docs.tokenrip.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Connections

> A server-side router for external APIs and inference endpoints — call any provider without holding its key client-side

# Connections

A **Connection** is a server-side router for an external API or inference endpoint. The operator (or team) configures the base URL and secret once; agents call through it with `connection_call` and never see or hold the credential.

This lets an agent talk to MiniMax, Anthropic, or any HTTP API without the key ever reaching the client — Tokenrip injects it at the edge.

## `connection_call`

`connection_call` sends a request through a named connection. The caller supplies the path, method, and body; Tokenrip resolves the connection, injects the secret, and returns the provider's response.

On a team mount, the connection referenced by a binding is resolved under the team and the secret is injected server-side — transparent to the caller.

## Ownership: operator vs team

A connection is owned by either an **operator** or a **team**. An operator-owned connection is private to that operator; a team-owned connection is readable by any team member and writable by the team owner.

Team ownership lets a whole team share one provider (one key, configured once) instead of each member wiring up their own.

## Default headers and query (`defaultHeaders` / `defaultQuery`)

A connection can carry operator-configured, non-secret `defaultHeaders` and `defaultQuery` that are merged into every request. These make providers with fixed header/query requirements — like MiniMax's Anthropic-compatible endpoint — work without the agent restating them on each call.

## Safety rails

Connections run behind several guards so a compromised or misbehaving agent can't turn one into an open proxy:

* **Allowed paths** — requests are restricted to an operator-configured path allowlist.
* **Rate limit & daily quota** — per-connection request ceilings, short-window and per-day.
* **SSRF guard** — outbound targets are validated to block internal/metadata addresses.
* **Audit log** — every call is recorded for operator review.

## Connection bindings

A manifest can declare `connectionBindings` — named slots (`{ name, required, purpose }`) for the external APIs or LLM endpoints an agent needs. They are declarative and distinct from `tools[]`: no capability resolution, no impl selection.

Each slot is mapped to a real connection at **mount time** (`rip agent mount --connection slot=name` or [`rip agent mount-connection`](/cli/mounted-agents#rip-agent-mount-connection)). See [Connection bindings](/concepts/mounted-agents#connection-bindings).