Create or replace the chat & workflow LLM quota for a user on this compute node

Writes the quota row for (computeNodeId, userId). Pass the literal string
__default__ as userId to set the node-wide default applied when no
per-user override exists.

Each cost-axis field is independent — omit (or set to null) any axis to
leave that axis uncapped. The per-workflow cap is forwarded to the LLM
governor as the per-execution budget when this user kicks off a workflow.

Permissions: Node owner or workspace administrator on public accounts.

Recent Requests
Log in to see full request history
TimeStatusUser Agent
Retrieving recent requests…
LoadingLoading…
Path Params
string
required

The UUID of the compute node

string
required

The Pennsieve user node id, or the literal string __default__ to
set the node-wide default.

Body Params

Cost ceilings applied to chat and workflow LLM spend for a single user on this
compute node. The compute node owner pays for LLM usage on their node, so they
set these limits to bound per-user spend.

Fallback resolution (axis-by-axis): any axis omitted (or set to null) on a
per-user row falls back to the node's __default__ row for that axis. Any axis
still unset there falls back to a small platform safety cap baked into
chat-service config — so unset axes are never "unlimited". Owners raise limits
by writing __default__ (node-wide) or by writing a per-user row with higher
numbers.

double | null

Maximum LLM spend (USD) per UTC day for this user on this node. Null/omitted
falls back to the node __default__ row, then to the platform safety cap.

double | null

Maximum LLM spend (USD) per UTC calendar month for this user on this node.
Null/omitted falls back to __default__, then to the platform safety cap.

double | null

Maximum LLM spend (USD) any single workflow invocation by this user may
incur. Forwarded to the LLM governor as the per-execution budget.
Null/omitted falls back to __default__, then to the platform safety cap.

string

Free-text notes (e.g. why this user has a custom limit).

Responses

Language
Credentials
Header
URL
LoadingLoading…
Response
Click Try It! to start a request and see the response here! Or choose an example:
application/json