[jamesbxwu]

After #704 and #696, the OpenAI template persists pending approvals in a bound Session Store, but the chat overlay guide, migration skill, and history reader docstring still describe them as process-local. Align these references with the existing harness implementation, including shared Runtime Store scheduling, accepted-approval recovery limits, and the fact that transcript history does not restore approval prompts.

This is the documentation follow-up to ML-70510; the runtime fix is already on main in #704 and #696 and is absent from the 0.5.1 release tag. Existing generated projects need the updated harness files; upgrading the distribution alone does not rewrite their source.

Validation: 156 scaffold/skill/CLI tests passed; 42 OpenAI SDK/checkpoint tests passed (1 live-model test skipped); 58 runtime/request-user tests passed (2 inapplicable LangGraph parameter cases skipped). Ruff and git diff --check passed. No live workspace resources were created.

CI: all checks pass except the six memory-backed test_history_orders_parallel_pending_messages variants in the framework job. The identical six failures already occur on the base commit a8d55f4 in main run 38662891535: expected pending-write order ["last", "first"], actual ["first", "last"]; both base and this PR report 6 failed, 1,994 passed, 24 skipped. This docs-only change does not touch that test or runtime path.

Bug-bash integration validation

The test-only combined branch at be2fa8ee8d6d3b3e011ecfa285e6455b90f9433a contains this exact PR head and eleven other reviewed heads. Its Agent Bricks unit suite passed 2,154 tests, 24 skipped. This is combined-source validation, not an isolated run of this PR; the branch must not be merged into main.

Private run 38698567494 passed all 20 live Claude cache tests at e231ca2 (OpenAI 8, LangGraph 4, generated adapters 8). Run 38701390693 at 53f9db8 passed OpenAI 8, LangGraph 4 and 3 adapter cases; the remaining 5 adapter cases returned HTTP403 because the shared test-workspace Gateway budget reached its $10 limit. Production source is identical across those runs and the final combined source; later changes affect only matrix fixtures, cleanup, regression tests and documentation. The rerun is not all green, and no budgets were changed.

The corrected Apps test passed 20/20 matrix rows, two grant snapshots, transitive-permission controls and repeat-deploy idempotency. Its evidence confirms clean source and wheel/source matching. That live run predates the final cleanup correction: the harness recorded twelve successful cleanup operations but omitted the implicit Runtime Store/role for direct-authoring Apps. #752 now includes ownership-checked cleanup for every App and safe retry/evidence handling, covered by focused regressions and the final unit suite. No new full model matrix was run for those cleanup-only changes.

Independent read-only run 38708044949 confirmed all 13 exact prior-run resources absent, including three consecutive typed404 observations for the omitted Runtime Store, dedicated database and role. One separate current-window database/role pair remains present and unattributed: SCIM did not recover its App identity and one bounded audit SELECT failed. That pair was left intact; full cleanup of the current matrix run is not established. No resources were deleted using name/time correlation alone.

The Apps matrix covers LangGraph, CLI/direct authoring, dev/deployed runtime and five tool kinds. Separate cache tests cover both LangGraph and OpenAI Agents. These checks do not establish fresh-account provisioning rights or freshly consented deployed Genie OBO acceptance.

All public checks finished. #694/#727 have 54 green; the other ten PRs have 53 green and the same framework job red. Its six memory-backed parallel-history ordering failures also occur on current main. No forced skips or weakened grant assertions were applied.