Ollama integration
Run large language models locally or in the cloud with Ollama
- Provider
- ollama
- Category
- Other
- Setup
- User-supplied API key
- Actions
- 8 available
Quick start
Call Ollama from your generated app's worker code via the OverSkill composio client. The toolkit slug below maps to Ollama's public Composio identifier — your app references it by name and OverSkill handles the OAuth round-trip.
// In your generated app's worker handler:
const composio = composioClient(env);
const result = await composio.execute(
"OLLAMA_ACTION_NAME",
{ /* action parameters — see Actions section below */ }
);
return new Response(JSON.stringify(result), {
headers: { "Content-Type": "application/json" }
});
Replace ACTION_NAME with one of the slugs listed in
the Actions section below. The composio client handles auth +
rate limits automatically — no API key is exposed to user code.
Setup
End-users supply their own Ollama API key. OverSkill provides a pre-built connection form via the integration card — your app can call the composio client immediately once the user has connected.
Available actions
8 actions exposed via the
Ollama integration. Each maps to a callable
slug — pass it to composio.execute(...) in your
worker code.
-
OLLAMA_CHAT -
OLLAMA_GENERATE -
OLLAMA_LIST_MODELS -
OLLAMA_OPEN_AI_CHAT_COMPLETIONS -
OLLAMA_OPEN_AI_COMPLETIONS -
OLLAMA_OPEN_AI_LIST_MODELS -
OLLAMA_SHOW -
OLLAMA_VERSION
Build with Ollama
Open OverSkill, describe what you want to build, and reference Ollama in your prompt — the AI will wire up the integration automatically.