Skip to main content
ask and chat run graph’s agent loop — the conversational surface for probing tools, prototyping workflows, and exercising plans. Every tool call here also feeds the shape cache, so exploration makes the planner smarter. Anything you’ll run twice belongs in a plan, and plans get their own surface: the workbench.

​
One-shot: graph ask

  • Streaming: the answer streams to stdout; tool activity (dim → tool {args} / ✓ tool 0.8s lines) goes to stderr. graph ask "…" > out.md captures just the answer.
  • Piped stdin becomes context:
  • --json buffers and emits a machine-readable envelope instead:
  • --no-stream prints the final answer only, without token streaming.
Put --thread after the message. It takes an optional value, so graph ask --thread "question" parses the question as a thread id.

​
Interactive: graph chat

Slash commands inside the REPL: A failed turn keeps the session alive and rolls the conversation back so a retry starts clean. On exit you get a copy-pasteable resume hint:

​
Reading the output

Dim text is progress; solid text is the answer. The distinction matters for piping — only the answer is on stdout.

Related topics

IntroductionThe shape cacheAsk stepsThreadsHow the agent picks tools