Claude Code routines: what they run, their limits, and when a hosted agent fits

What the docs say a routine does, its caps and limits, and where a hosted agent fits next to it.

Plori

What does a Claude Code routine run, where does it stop, and when does a hosted agent fit next to it? Every statement about routines comes from Anthropic's documentation as we read it on September 24, 2026. We did not run a routine for this post, and Plori sells an alternative for some of this work.

The short answer

  • A routine is a saved prompt, one or more GitHub repositories and a set of connectors, run in the cloud on a schedule, an HTTP call or a GitHub event. Routines are in research preview.
  • Use a routine for repository work that ends in a branch or a pull request. It uses your subscription usage and a daily run cap.
  • Consider a hosted agent next to it when the work has no repository, needs files that stay between runs, or must bill apart from your Claude plan.

Limits listed in the docs

Limit Value
Shortest schedule interval 1 hour
Daily run cap Per account, in addition to usage limits
GitHub events Hourly caps during the preview
Sharing Not shared with teammates
/schedule Needs a claude.ai subscription login

The docs do not print the daily cap. Anthropic's launch post of April 14, 2026 gave 5 runs per day on Pro, 15 on Max, and 25 on Team and Enterprise.

When it applies

  1. A run has no permission-mode picker. It uses every tool from an included connector, including writes, without asking. A green status does not mean the task in your prompt succeeded, so read the transcript.
  2. Each run clones the repositories again and starts a new session. The docs do not describe a working directory that stays from one run to the next.
  3. A routine is the better choice when you already pay for a Claude plan and stay under its limits, or when the task needs the connectors on your claude.ai account.

The full post has the headless mode comparison, a six-step docs-drift setup, and how to call a Plori agent from Claude Code over MCP.

Read the full post

Try plori.ai

Subscribe to Plori Engineering

Engineering notes building AI agents in the cloud

Free. Unsubscribe anytime.

Keep reading

All issues →
  1. Scion-style merges: one comparison takes 1.93 s at 100,000 files

    What we took from Anthropic's Scion for Workspaces, and the merge cost we measured before changing anything.

  2. Claude Code handed 0 of 240 unnamed-server prompts to our MCP server, then 21

    What Claude Code shows the model about MCP tools, and what a rewrite of our server text changed in 240 runs.

  3. AI agent files: a local disk, saved to object storage every 60 seconds

    What a crash can lose, how results are accepted, and the start and save times we measured.