📖About

First increment of recursive-self-improvement for the BUILD PIPELINE. Owner: Elrond (the build-pipeline team owns the whole loop incl. prioritize + guardrails; Clara/Themis are not on this team). v1 ships ONE metric and its dashboard; metrics are added incrementally later. METRIC: success rate as 'successes out of 10' per execution mode (Orion bypass, Albus bypass, planner); target 10/10. Success signal by mode: Orion/Albus bypass = a bypass session that ended completed (vs failed), read from each owner's bypass book session events; planner = an initiative review verdict 'pass' (vs not) on planner-mode initiatives. ARCHITECTURE (agent-centric): computation lives in an Elrond droid (agents do the work); the dashboard is a thin observability window that calls it via a shim (design 173). v1 computes live; the persistent scoreboard store + scheduled collector droid are the NEXT increment. DASHBOARD: weekly average success-out-of-10 over the recent 10 rolling weeks, with a filter for the 3 modes (All/Orion/Albus/Planner) and a target line at 10. Built via Orion bypass.

⚖️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)
• v1 ships ONE metric: build success-out-of-10 per mode, live-computed by Elronds Pipeline Success Metric droid; the dashboard is a thin window calling it via a shim. — Agent-centric (design 173); collector+scoreboard store deferred to next increment per incremental approach. (Orion (or-bx-01))
• Elrond owns the whole RSI loop including prioritize and guardrails (Clara/Themis are not on the build-pipeline team). — Per Ninad correction. (Orion (or-bx-01))
• Dashboard live on 7260, registered (app pipeline-rsi + droid el-psm-01, owner Elrond), routed at /dev/pipeline-rsi/ via Radagast; weekly success-out-of-10, 10 rolling weeks, mode filter, target line 10. — Orion bypass build. (Orion (or-bx-01))
• [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

• Dashboard shows weekly success-out-of-10 over the recent 10 rolling weeks per execution mode, with a filter (All / Orion bypass / Albus bypass / planner) and a target line at 10. (must_have)
• The success metric is computed by an Elrond droid (not in the app); the dashboard is a thin window calling it via a shim. Owner = Elrond. (must_have)
• Per-mode success is correct: Orion/Albus bypass from bypass-session completed-vs-failed; planner from review verdict pass-vs-not on planner-mode initiatives. (must_have)