docs: clarify which standing approvals repeat work

[santoshkumarradha]

Supplemental correction verified in separate PR #1612

Separate PR #1612 adds task-aware one-off wording: Run it then / Runs the work then, while retaining reminder-specific text. The live recording shows the corrected work card and its decline.

Tested revision f855f2e6510c155b282cba388db2f92d2121dda3 on Spark: a practical shopping-note workflow, 19 generated-project tests, unchanged inventory and zero saved recurring definitions. All 30 receipts used OpenRouter deepseek/deepseek-v4.1-flash; Model Pool was disabled as disclosed in coverage/limitations.

Fix PR #1612 · Video · GIF · Original terminal recording · Actual corrected card · Post-approval execution evidence

This evidence tests the separate corrective revision; it does not claim this PR’s original head already contains that fix. Earlier recordings and their limitations remain below. Tomorrow’s task was declined, not executed; ordinary model-action and permission limits still apply. Integration boundary: this #1612 base does not contain #1560 does.isolate, so ordinary Once tools do not prove isolated execution. Consolidation must suppress Once for isolated tasks or use a shared isolated executor. Isolated Once and watch/probe Once were not live-verified.

Real product workflow — recorded on Spark

Goal: Prepare a practical pantry shopping plan and choose the right ongoing commitments: a one-off reminder, a weekly reminder and a standing permission rule.

Observed outcome: The app wrote shopping-plan.md. The one-off reminder fired and retired; the weekly reminder was approved then stopped through the product; the permission rule remained active without a wake. The cards and stored states distinguish one-off, repeating and always-on commitments.

Watch the workflow (MP4) · Animated GIF · Original terminal recording (.cast)

Evidence: tested revision 82e34d5dcca434536398a08c0a2849c49a4f6ae0; Spark session critical-extended-1520-v2. Execution/binary identity, model receipts, checksums and playback details. All 34 recorded calls used OpenRouter deepseek/deepseek-v4.1-flash, including auxiliary calls. Final saved standing states.

Playback and limits: Original .cast is preserved; GIF/MP4 play at 3× speed with idle intervals capped at 3 seconds. Run used --one-model and explicitly disabled Model Pool for issue #1608; Model Pool behavior is verified separately in #1610. Earlier failed-policy recordings remain preserved locally and are not counted as successful evidence. The model briefly described an already-approved weekly reminder as pending; a user correction and listing resolved it. The recording keeps that rough edge. The PR is documentation-only; this demonstrates the documented current product behavior, not a new implementation of reminder labels.

Explain the different engine-owned consequences of accepting a reminder, rule or repeating check. The manual quotes the labels and consequences the current cards actually display.

Related to #1505. This PR changes documentation only; the current engine already gives one-off reminders a non-repeating consequence.

Validation: live tmux with OpenRouter deepseek/deepseek-v4.1-flash proposed a one-off reminder and displayed Reminds you then. Nothing repeats. The proposal was declined after inspection. All auxiliary-call receipts named the same model. GitHub light gate, touched packages, aggregate check and CLA all passed.

[santoshkumarradha]

@AbirAbbas this is review ready

[santoshkumarradha]

Superseded by draft #1627. Exact reviewed head 82e34d5dcca434536398a08c0a2849c49a4f6ae0 is included in published Santosh/dev history (integration b096a949c8b349a9b75d2e751783564d02890288), with remote ancestry verified and make build passing after the merge. The aggregate preserves links to this PR’s original live recording, screenshots, model receipts, checks and limitations. Closing this source PR as replaced; its branch and review/media history are retained. This does not merge it into dev, and the aggregate remains draft.