Skip to content
Agentsmedium signalverified

GitHub sets three Copilot dates: seats paid upfront, one chat policy, longer retention

One changelog entry on August 28 carries three separate changes. From September 1, new Copilot Business and Enterprise seats bought by card or PayPal must be paid for before anyone can use them, and existing card customers move to upfront charges on October 1. No earlier than September 28, chat on github.com, chat in GitHub Mobile and the cloud agent become one experience under one policy that is on by default, and chat data then stays for the life of the account instead of 28 days. On the same date, the default effort level for Copilot code review moves from Lite to Balanced.

By Redakcija WebAiRadarPublished 2 min readwritten by a model
Image: GitHub

GitHub published three Copilot changes in a single changelog entry on August 28. Two of them are billing and packaging, and read like housekeeping. The third is a retention change that arrives inside the second one, costs nothing, and is enabled by default.

Seats are paid before they are used

From September 1, 2026, GitHub reopens sign-ups for new Copilot Business and Copilot Enterprise customers who pay by credit card or PayPal, with what it describes as stronger account vetting. Every new seat assignment on those plans requires payment for that seat before the user gains access, and at the start of the next billing cycle all assigned seats incur an upfront charge. Existing customers paying by card or PayPal move to the same arrangement on October 1, 2026.

Prices are not changing. What changes is when the money moves and what happens when you undo a decision: revoking a seat produces no prorated refund, and the removal shows up in the next monthly cycle. Seats added mid-cycle are still prorated from the day they are assigned, included usage may be prorated across the month to match that, and spend controls, usage tracking and the option to buy additional AI credits all remain.

Three surfaces, one policy, and a retention change inside it

No earlier than September 28, 2026, Copilot Chat on github.com, Copilot Chat in GitHub Mobile and the Copilot cloud agent relaunch as a single experience governed by a single policy, which is enabled by default after launch. The cloud agent moves onto Sandbox, and Copilot on github.com migrates fully to the agent sessions experience that until now only the cloud agent used.

That migration carries the part worth reading twice: chat data will be retained for the life of the account instead of 28 days. Opting out does not preserve the old behavior — teams that opt out lose Copilot on github.com and in GitHub Mobile once the new experience launches. Business and enterprise administrators set this in Copilot settings on github.com, under the Copilot cloud agent entry, and GitHub asks them to review it before September 28.

The default review effort moves up

Also on September 28, 2026, the review effort value named Default starts to mean Balanced rather than Lite, for existing and new repositories and organizations using Copilot code review. An organization default applies to every repository that has not chosen its own level, and a repository default applies to automatically requested reviews. To keep the lighter analysis, you have to select Lite explicitly at the repository or organization level before that date; GitHub says an explicit choice will be respected. For manually requested reviews the level is still picked in the Reviewers bar of the pull request.

BrandsCopilot

Related

OPENCLAW 2026.10.232,768characters, and past that a key could slip thr
Agentsmedium signal

OpenClaw 2026.10.2 stops long tool output from slipping credentials past redaction

A bug in OpenClaw's secret masking let a credential through whenever it straddled the boundary between two slices of a long text. The fix ships in 2026.10.2 along with repaired updates from 2026.10.1 and a new compaction default for OpenAI and Anthropic API keys.

OpenClawverified

The OpenClaw web interface with a conversation on the left and the Bits & Bolts reference MCP app open beside it, showing a 3D keycap model and an Attach view + note button.
Agentsmedium signal

OpenClaw 2026.10.1 lets agents search installed skills and changes how sessions message each other

OpenClaw released version 2026.10.1 on October 10, 2026, its first stable release since 2026.9.7. Agents can now search the skills installed on a machine and read only the one a task needs, instead of carrying the whole library in the prompt. Opt-in MCP Apps run beside chat, and full backups can go to an external drive or Cloudflare R2. Three changes can break existing setups: session-to-session messaging, the old required argument, and Workboard data stored before July 2026.

OpenClawverified

CLAUDE CODE2.1.294hooks written as instructions now block what t
Agentsmedium signal

Claude Code 2.1.294 fixes prompt hooks that let through the actions they were written to block

Anthropic released Claude Code 2.1.294 on October 8, 2026. It fixes prompt and agent hooks phrased as instructions, which could allow exactly the actions they were meant to stop. Version 2.1.295 followed the same day with an onFailure setting that makes a crashed or timed-out command or HTTP hook block the action. Releases 2.1.292 and 2.1.293 from earlier in the week closed several permission and sandbox gaps.

Claude Code Releases (GitHub)verified