📖About

Ninad: extend Elrond's RSI Improvement Engine to (1) run EVERY HOUR (was daily) and (2) add a per-blocked-initiative diagnosis pass that diagnoses ONLY blocked initiatives that were never diagnosed by the RSI engine OR got re-blocked AFTER their last RSI diagnosis (skip ones already diagnosed and not since re-blocked), then files+dispatches an improvement initiative for each. Tracking via a 'RSI diagnosed:' decision marker on the blocked initiative; selection compares the initiative's last '-> blocked' decision time vs its last 'RSI diagnosed:' marker time. Idempotent: a continuously-blocked initiative is diagnosed once; only a genuine new block after diagnosis re-triggers. Dedup: skip if an open improvement initiative (tag rsi-blocked:) already exists. Keep the existing lane-scoring pass. Change the scheduler cadence (Hermione's scheduled_jobs row for el-rsi-eng-01) from daily to hourly.

⚖️Decisions

• 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)
• Extended Elrond's RSI Improvement Engine (per Ninad's choice over Albus's diagnostician). Added an hourly per-blocked-initiative diagnosis pass: selects currently-blocked initiatives that were NEVER diagnosed by the RSI engine OR re-blocked after their last RSI diagnosis (compares the initiative's last '-> blocked' decision time vs its last 'RSI diagnosed:' marker time); for each, runs an LLM root-cause diagnosis, files+dispatches an improvement initiative through the Author, tags it pipeline-rsi + rsi-blocked:<id>, and records an 'RSI diagnosed:' marker on the parent for idempotency. Dedup: skips if an open improvement (rsi-blocked:<id>) already exists. A continuously-blocked initiative is diagnosed once; only a genuine new block after diagnosis re-triggers. Kept the existing lane-scoring pass. Schedule changed from daily (cron 45 2) to hourly (cron 0 * * * *). Dry-run verified clean (0 blocked currently). Runs as a fresh cron subprocess so new code loads next run. (orion)
• [status-router] executing -> reviewing | event=bypass_reviewing | bypass transition (or-bx-01)
• [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

• The RSI engine runs hourly (scheduler cadence changed from daily to ~3600s) (must_have)
• Each run diagnoses ONLY currently-blocked initiatives never diagnosed by the RSI engine, or re-blocked after their last RSI diagnosis; already-diagnosed-and-not-re-blocked ones are skipped (must_have)
• For each selected blocked initiative it diagnoses the root cause and files+dispatches an improvement initiative, records a 'RSI diagnosed:' marker, and skips if an open improvement (rsi-blocked:<id> tag) already exists (must_have)