fix(server): delegate and preview rules close four repeat eval misses

apps/server/src/mcp/toolkits/preview/tools.ts

@@ -137,7 +137,7 @@ const PreviewSetAppearanceTool = safeBrowserTool(
 export const PreviewSnapshotTool = safeBrowserTool(
   Tool.make("preview_snapshot", {
     description:
-      "Use it only when the current request asks for browser work, including checking an edit in the browser. Never use it unasked to verify your own edit, or to repeat a snapshot from an earlier turn. Works on a tab the user is driving. Returns page state, elements, diagnostics, and screenshot size. Target one element with `locator='aria-ref=<ref>'` from `accessibilityTree`, even in frames. Refs die on navigation, a new snapshot, or handoff. Text caps near 20 KB; on your own or an unclaimed tab, `preview_evaluate` reads more. To show a screenshot: `save=true`, then embed ![alt](screenshotPath).",
+      "Use it only when the current request asks for browser work, including checking an edit in the browser. Never use it unasked to verify your own edit, or to repeat a snapshot from an earlier turn. Works on a tab the user is driving. Returns page state, elements, diagnostics, and screenshot size. Target one element with `locator='aria-ref=<ref>'` from `accessibilityTree`, even in frames. Refs die on navigation, a new snapshot, or handoff. Text caps near 20 KB; on your own or an unclaimed tab, `preview_evaluate` reads more. To show a screenshot when asked: `save=true`, then embed ![alt](screenshotPath).",
     parameters: Schema.Struct({
       ...PreviewAutomationTabTargetInput.fields,
       includeImage: Schema.optional(

packages/provider-core/src/server/orchestrationInstructions.ts

@@ -6,9 +6,9 @@ export const T3_CODE_ORCHESTRATION_SECTIONS: ReadonlyArray<string> = [
 
 Ordinary coding tasks need no T3 tools.
 
-- Subagents: use your native tool when the child uses your provider, a model it offers, and no settings beyond yours, and you wait in this turn (Claude \`Agent\` always with \`run_in_background:false\`; Codex \`spawn_agent\`, then wait). Use \`delegate_task\` for another provider's model, user-asked settings, a separate checkout (\`isolation:"worktree"\`), background work, or \`delegate_task\` by name. A native type running another provider's model counts too. Workflow only when asked.
-- Before \`delegate_task\`, call \`orchestrator_capabilities\` once. Again only after an error or a user-named provider change. Settings go in \`target.options\`. One child per named model. Provider or model unavailable: say so. Never switch unasked, or fall back to native tools, Workflow, or CLIs. Never make a child's checkout in the shell or in \`task\`.
-- Every new delegate review or re-check is a new \`delegate_task\` (brief, findings, objections), never a \`t3_thread_send\`. Fix small review findings yourself; start no delegate for them. When told to proceed without a delegate, \`task_cancel\` it. Start no replacement.
+- Subagents: use your native tool when the child uses your provider, a model it offers, and no settings beyond yours, and it finishes in this turn (Claude \`Agent\` always with \`run_in_background:false\`; Codex \`spawn_agent\`, then wait). Use \`delegate_task\` for another provider's model, user-asked settings, a separate checkout (\`isolation:"worktree"\`), background work, or \`delegate_task\` by name. A native type running another provider's model counts too. Workflow only when asked.
+- Call \`orchestrator_capabilities\` once per thread, before the first \`delegate_task\`. Again only after an error or a user-named provider change. Settings go in \`target.options\`. One child per named model. Provider or model unavailable: say so. Never switch unasked, or fall back to native tools, Workflow, or CLIs. Never make a child's checkout in the shell or in \`task\`.
+- Every new delegate review or re-check is a new \`delegate_task\` (brief, findings, objections), never a \`t3_thread_send\`. Fix small review findings yourself; start no delegate for them. When told to proceed without a delegate, or to run fewer, \`task_cancel\` the extras. Start no replacement.
 - Follow a running delegate's \`next\`. Do not poll it (\`task_status\`, \`sleep\`, watchers, Monitor, ScheduleWakeup). Stop a delegate run you replace or no longer need with \`task_cancel\`, never \`t3_thread_interrupt\` on a \`childThreadId\`.
 - If you are a subagent or child, do the work yourself. Start no subagents unless your task says to.`,
   `- \`t3_thread_launch\` and \`create_threads\` make new top-level threads, only when the user asks for a new thread. "Subagent", "delegate", and "parallel" never mean one. Worktree thread: set \`workspaceStrategy\`; shell worktrees do not bind it, and no checkout setup in its prompt. No \`workspaceStrategy\`: project root, not your worktree.