3 changes since v2.1.311, landing in 5 of 23 prompts.
Every prompt v2.1.312 sent, compared word for word with the capture before it. Each card is one edit, with every prompt it reached; the table after them shows where each one landed.
What changed 3
1
System prompt · under “Context management”
EndConversation deferred-tool guidance removed
The system prompt no longer mentions the deferred EndConversation tool. It previously said to use it only for sustained user abuse aimed at the assistant or when the user explicitly asked for a demonstration, and to load its full guidance via ToolSearch first. This appeared in only 1 of 23 captures, so it likely applies to one specific setup.
Behind a feature flag
+0−34 words
−EndConversation (deferred tool): use only for sustained user abuse directed at the assistant, or when the user explicitly asks to see it demonstrated. Load the full guidance via ToolSearch("select:EndConversation") before using it.
2
AskUserQuestion tool description · line 12
AskUserQuestion questions must make sense without earlier context
The tool description now warns that the user may see only the question dialog, without tool results or earlier text from the turn. Claude is told to make each question self-contained, briefly state any findings it relies on, and name the specific items being asked about. This addition appeared in 2 of 5 captures, so it may apply only to some setups.
+57−0 words
+The user may see only the question dialog: tool results and text you wrote earlier in this turn may not be on their screen while it is open. Make each question stand on its own: if it depends on something you found, say so briefly, and name the items you are asking about in the question itself.
3
Appended context (skills, environment) · at “EndConversation”
EndConversation dropped from appended tool listing
The name EndConversation was removed from the appended context, so it is no longer listed there as an available deferred tool. This was seen in only 1 of 23 captures and matches the removal of the EndConversation guidance from the system prompt.
Behind a feature flag
+0−1 words
−EndConversation
Where each change landed
Every prompt the release sent, one row each. A number is a change from the list above; click it to jump there. Rows with more than one model string count them, because each model gets its own copy of the prompt.
18 prompts did not move at all: Agent SDK, default, Agent SDK, seeded with this account's feature flags, Agent SDK, seeded with anonymous feature flags, Auto-mode classifier, Slash commands: /init, Slash commands: /security-review, Slash commands: /team-onboarding, System reminders: Claude md, System reminders: Hook pretool block, System reminders: Read dup, System reminders: Read empty, System reminders: Read offset, System reminders: Skill run, System reminders: Todo nag, System reminders: Toolsearch deferred, Subagents: explore, Subagents: general-purpose, Subagents: plan.
Set aside as capture noise
These change on every capture by construction, so they are masked before comparing and never counted as a change.
