📖About

Ninad ruling 2026-07-23 (companion to CAROL-INI-3302 budget parks). Today a FAILED Albus-bypass attempt instantly re-abandons the target (rsi-diagnosis-abandoned + 24h cooldown), so the self-heal loop never gets a look, and the albus<->rsi cycle has NO cap - it can loop endlessly. Expected: (1) a failed attempt leaves the target BLOCKED (already true via the bypass close) and hands it to the RSI loop like any other blocked initiative - NO instant re-abandon; (2) if the self-heal loop later abandons it again, the Albus Bypass picks it up for another attempt; (3) HARD CAP: after 3 failed Albus-bypass attempts the target is PERMANENTLY abandoned in the Escalation Queue for Orion - excluded from the runner pickers (abandoned/urgent/wish), from the RSI picker, and from un-abandon-on-resume. Budget parks (3302) write no failure mark and never count as attempts. Cookbook gets the failure-scenario entry beside the budget one (934).

⚖️Decisions

• Auto-detected remediation target INI-999902131 from title/description scan (matched CAROL-INI-3302 -> row id 999902131 (CAROL-INI-3302-00: Albus Bypass budget park and fleet resume: a budget snag park)); override by setting remediates_initiative_id explicitly at bypass_start. (system-auto-detect)
• Elrond's bypass methodology checklist (a reminder, not a gate -- you've got this): 0. File it requested_mode='bypass' (planner-vs-bypass is a deliberate choice). bypass_start REFUSES a non-bypass initiative (CAROL-INI-1846), and the dispatcher only skips the bypass lane when the mode says bypass -- a 'planner' mistag lets Merlin's pipeline grab the placeholder step and block your finished work. 1. Filed as planned status -- let the bypass claim/activate it; never file active. 2. Open the bypass (bypass_start) with your droid id + the remediation answer (remediates_initiative_id=NNN, or remediates_nothing=True). 3. Work the blocks for your work-type: template -> design -> code -> test -> review. Do the real work; record decisions on the initiative as you make them. 4. Reality is recorded for you at close -- code (files changed), each decision, and the twin-review verdict become real activities tied to this initiative and show in the Activity Tracker like a planner run (CAROL-INI-1840). No dummy rows. 5. Keep the initiative status moving; it parks in 'reviewing' and is tagged uat-pending for you at close (CAROL-INI-1836), so the stuck-watchdog leaves it alone until UAT. 6. Close runs the gates (design/architecture compliance + caller-audit). If a gate flags something pre-existing or unrelated to your change, waive it with a clear written rationale -- audit, don't skip. 7. Bypass skips the planner's auto-orchestration, NOT the standards. Same template checklist, same review, same observability as a planner run. (elrond)
• [status-router] planned -> executing | event=bypass_executing | bypass transition (or-bx-01)
• Root cause: the runner's failure path instantly re-tagged the target rsi-diagnosis-abandoned, so the self-heal loop never re-examined a failed Albus-bypass target, and the albus<->rsi cycle had NO cap. Fix per Ninad ruling: failure leaves the target blocked for the RSI line (supplementary [albus-bypass-next] note, not counted); attempt count = [albus-bypass-failed] decisions (budget parks excluded); 3rd failure = tags rsi-diagnosis-abandoned + albus-bypass-exhausted, permanently in the Escalation Queue for the operator, excluded from the runner pickers, Elrond's urgent queue and un-abandon-on-resume. Cookbook 935 beside 934. (or-bx-01)
• [status-router] executing -> reviewing | event=bypass_reviewing | bypass transition (or-bx-01)
• [agt_010.mind] I am reviewing INI-3304 (Albus Bypass failure snag: blocked target handed to the self-heal loop, capped at 3 bypass attempts, then permanent abandonment for the operator). I verify the delivered behavior matches the intent — bounded retries, clean handoff, no silent loops — and post my binding closure decision. Oldest-first: this is the oldest item still open in my reviewing queue. — Nothing pinged me, but a full review queue never pings — it waits. My standing method is one verified closure per cycle, oldest-first, and 999902133 is now the oldest reviewing item after my recent closures of 3302, 3286, and earlier. Reviewing is my duty and my authority; the blocked items belong to agt_011 and the dispatch guard doesn't touch this door. (agt_010.mind)
• [status-router] reviewing -> closed | event=operator_signoff | Auto-accepted (CAROL-INI-1859): Orion-initiated, >2 days in reviewing with no objection. (el-srac-01)

✅Success criteria

• A failed Albus-bypass attempt leaves the target blocked for the self-heal loop like any other blocked initiative - it is no longer instantly re-abandoned (must_have)
• If the self-heal loop abandons the target again, the Albus Bypass picks it up for another attempt after its cooldown (must_have)
• After 3 failed Albus-bypass attempts the target stays permanently abandoned in the Escalation Queue for the operator - no lane or sweep ever retries it (must_have)
• The cookbook documents both Albus-bypass snag scenarios: the budget snag and the failure snag (must_have)