Anthropic adds monthly credit for Claude Agent SDK
Anthropic adds a dedicated monthly credit for the Claude Agent SDK to Pro, Max, Team, and Enterprise plans beginning June 15, 2026. The credit covers custom scripts, claude -p commands, GitHub Actions integrations, and third-party apps. Interactive feature limits remain unchanged. API key accounts and some Enterprise seats receive no credit and stay on pay-as-you-go billing. Credits refresh each cycle without rollover.
This is misleading.
This policy redefines the term "interactive" to mean "using an Anthropic front-end".
If you use `claude -p` or Agent SDK to do something interactively, it now uses credits, not your subscription limits. So the "interactive use" heading saying "unchanged" subscriptions is not accurate.
Starting June 15, paid Claude plans include a monthly Claude Agent SDK credit.
It covers usage on your own scripts and agents, claude -p, and third-party apps built on the SDK (OpenClaw, Conductor, etc) and it's separate from your regular usage limits.
Starting June 15, paid Claude plans include a monthly Claude Agent SDK credit. It covers usage on your own scripts and agents, claude -p, and third-party apps built on the SDK (OpenClaw, Conductor, etc) and it's separate from your regular usage limits.
The comment section tells you everything.
I mostly use Claude Agent SDK (~80%) and sometimes Claude Code interactively (~20%).
I prefer my own harness/UI over Claude Code CLI/Cowork. Most of my use cases with agents involve programmatic use (e.g., long-running loops and automations).
Enabling devs to build and work with their own harnesses should be encouraged. That's not the message I am getting here.
I appreciate the credits, but only time (when this comes into effect) will tell how bad it is and how it affects my use cases and overall usage. I hate that uncertainty in these times.
I do understand that this decision helps clarify usage, but it's obviously going to affect how much I can leverage the subscription itself.
Glad I decided to move a lot of my work to Codex over the past couple of weeks, where I get to freely decide how I use my subscription. We need more of this in the space.
Starting June 15, paid Claude plans can claim a dedicated monthly credit for programmatic usage. The credit covers usage of: - Claude Agent SDK - claude -p - Claude Code GitHub Actions - Third-party apps built on the Agent SDK
I appreciate the clarity. But this feels like it will negatively affect folks using that subscription heavily for programmatic use cases. I wondered if the Anthropic team actually considered that. Or maybe there is something I am not understanding.
I agree with this. I think it's a huge mistake if the goal was to become "the foundational infrastructure for agents." There are so many great developers (probably the best agent engineers I know) building on top of the Claude Agent SDK who are starting to move away from the Claude ecosystem because of these decisions.
I appreciate the clarity. But this feels like it will negatively affect folks using that subscription heavily for programmatic use cases. I wondered if the Anthropic team actually considered that. Or maybe there is something I am not understanding.
their attitude is downstream of their compute constraint. but if the API is the way to go i think there will be a real market for model substitution.
i was fighting with the harness and model for a lot of my last claude time if we're being honest. it's gotten better but i don't fully forgive it idk
Shrinkflation
Anthropic is trying very, very hard to have maximum price discrimination based on whether or not your code is being run “programmatically” or not.
This, not surprisingly, is kind of confusing to programmers.
Starting June 15, paid Claude plans can claim a dedicated monthly credit for programmatic usage. The credit covers usage of: - Claude Agent SDK - claude -p - Claude Code GitHub Actions - Third-party apps built on the Agent SDK
Also not surprisingly there is already a workaround
hey surprise - you can just launch interactive in tmux and then tail the jsonl - shipped a small wrapper...ralph loop iterating to full parity rn https://github.com/dexhorthy/shannon
It’s dramatically more expensive but you get a few free credits toward the new expense to soften the blow.
Your users aren’t dumb. Stop trying to dress everything up in 10 layers of PR speak and just say what it is.
Somewhat tired of playing 'cat and mouse' games with ever-changing Claude Code rate limits 😼
The easiest way to overcome the new -p changes (starting mid June) is to run an interactive session in a multiplexer like tmux and to pipe commands/outputs:
headless claude -p "<do stuff>" --tmux headless --list headless send headless-claude-<id> -p "continue."
👉 https://github.com/RobertTLange/headless-cli/
I’ll admit it’s nice seeing T3 Code called out in official Anthropic comms
If you use any of the following with your Claude sub, your usage must got cut by 25x: - T3 Code - Conductor - zed - jean - “Claude -p” in your ci - scripts to call Claude code from other tools
They’re disguising this as “free credits”. Don’t fall for it.
Starting June 15, paid Claude plans can claim a dedicated monthly credit for programmatic usage. The credit covers usage of: - Claude Agent SDK - claude -p - Claude Code GitHub Actions - Third-party apps built on the Agent SDK
guess how many turns $20 in API credits last.