OpenRouter chat completion brief
Read-only OpenRouter workflow for creating one explicit non-streaming chat completion and summarizing answer, route metadata, token usage, and cost. It never enables streaming, tools, plugins, files, media generation, presets, guardrails, key management, workspace administration, or arbitrary OpenRouter routes.
How it works
Inspect the data fetches, transforms, gates, and output this recipe runs.
Process (6 steps)
completion
?
choices
default
Apply default
choice_count
count
Count items in choices
choice_table
to_table
Format results as a data table
summarize
?
card
to_summary
Format results as a summary card
Settings
Configurable at install. Defaults shown — change them anytime in Recued.
focus
setting
=
Summarize the OpenRouter completion, selected model, token and cost usage, routing metadata, and any practical follow-up. Do not invent facts outside the API response.
model
setting
=
openai/gpt-4
top p
setting
=
1
prompt
setting
=
metadata
setting
=
[object Object]
max length
setting
=
900
max tokens
setting
=
900
openrouter
setting
=
session id
setting
=
temperature
setting
=
0.2
system prompt
setting
=
You are a concise assistant. Answer directly, state uncertainty, and do not invent facts.
reasoning effort
setting
=
none
provider preferences
setting
=
[object Object]
Trust & control
What installing this recipe would let it do. Recued grants these permissions at install — review them there before approving.