Sign in
← AI PulseSep 1, 2026

Wire · news · Single-source brief

Anthropic Details Claude API Sessions Endpoints

Anthropic's API reference documentation outlines the structure and functionality of Sessions endpoints for Claude models, including advisor configurations and cost management features.

By Illumora Editorial

Source · Sep 1, 2026, 5:55 PM · On Illumora · Sep 1, 2026, 5:58 PM

Media from the primary source — shown here so you can stay on Illumora.

Rewritten from one allowlisted primary — not independent enterprise reporting. Lanes →

Brief drafted by Illumora’s editorial model from the linked primary source. Ops desk reviews flagged pieces. How we write →

Read the source →Anthropic Docs (prompting) — Sessions - Claude API Reference
Save

Anthropic has provided API reference documentation for its Sessions endpoints, detailing how users can interact with Claude models within a session context. This documentation covers various aspects of session management, from configuring advisor models to setting financial controls.

Key Points

  • A platform advisor roster entry allows a session's primary thread to consult a Claude model mid-turn.
  • The anthropic.advisor roster name is reserved for this advisor entry.
  • A specific Claude model ID must be permitted as an advisor for the agent's model.
  • Users can specify a hard spend ceiling, max_list_cost, which stops new model requests once the tracked list cost is reached.
  • Monetary amounts for max_list_cost are specified in minor units of the currency, as an integer decimal string.
  • USD is the only currency currently supported for cost tracking.
  • Prompt-cache creation token usage is broken down by cache lifetime, including 1-hour and 5-minute ephemeral cache entries.

Context

According to Anthropic's documentation, the Sessions API allows for detailed control over how Claude models operate within a conversational or task-oriented session. This includes the ability to integrate secondary Claude models as advisors, which can be consulted during ongoing interactions. The documentation also highlights mechanisms for managing the financial aspects of API usage, such as setting expenditure limits and tracking token consumption for caching.

Why It Matters

Builders can use these API specifications to design more sophisticated and cost-controlled applications leveraging Claude models. The ability to configure advisor models enables complex multi-agent interactions, while cost ceilings provide a critical tool for managing operational expenses and preventing unexpected overages.

What To Do

  • Review the sessions/threads/advisor specification to understand how to permit advisor models.
  • Note the max_list_cost parameter for implementing hard spend ceilings in session configurations.
  • Observe the format for specifying monetary amounts in minor units and the current USD-only currency support.
  • Examine the prompt-cache creation token usage details for 1-hour and 5-minute ephemeral cache entries to optimize caching strategies.

Keep Exploring

/atlas/claude-family