Apply plan-review fixes within the operator-authorized scope, preserving plan versions. Prepare concrete diffs for any unresolved scope or decision; prior authorization remains valid and silence never grants approval.
Ultra-compressed communication mode. Cuts output tokens 65% (measured) by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite (this repo default), full, ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says…
Post-build module quality review. Checks plan adherence, linguistic accuracy, pedagogical quality, activities, vocabulary completeness, engagement.
Audit prompt sizes across build orchestration dirs. Surfaces sections that dominate the prompt (>40%), tracks per-phase budget vs the model's context window, and checks whether directives in the prompt were actually covered in the response.
Model-agnostic orchestrator playbook for driving ONE epic/track lane end-to-end over the fleet-comms message plane. Invoke this when you are launched as an epic/track driver (Grok, Sonnet-5, Gemini/AGY, Kimi, Claude, or Cursor) via a start-<model>-driver.sh…
Inspect or reconcile Ukrainian dataset work under the operator's human-authored-source requirement, preserving rights, provenance, custody, held-out separation, and independent review. Legacy V4 row generation is stopped and this skill cannot restart it.
Run bounded body-free context-link intake for non-trivial work, provenance questions, and handoffs (ADR-018). Use relevant private recall only when it materially restores task continuity. GitHub, Fleet Comms, Monitor, session streams, rollover, and formal…
Canonical code/infra closeout workflow — freezes scope, resolves the exact review target (local/commit/branch/PR), runs a non-mutating review, resolves a cross-family reviewer, and requires separate behavior proof for user-visible code or infrastructure…