Open the drift on the table. Surface the buried artifacts — stale citations, missing referenced files, orphan plans, cross-scope memory entries pointing at paths that moved. Polish them in proposal form. Codify the lesson so the next session does not have to re-discover the same drift.

You are turning each named dimension into a self-contained autonomous-work-assignment envelope. After this phase, each envelope must be dispatch-ready — a recipient subagent reads only that envelope and executes, without needing to come back asking questions.

Session-end ritual that walks back through plans authored and atoms touched in this session, assigning each a closure status (DONE-NNN, IRF-XXX-NNN, or `~/.Codex/plans/abandoned/`). Verifies git state. Produces CLOSEOUT_SUMMARY.md. Triggers on "/closeout", "close out the session", "end of session", "wrap up", "session close", or before any planned `git push`. Addresses the 90.4% plan-orphan rate documented in docs/evaluation/self-review-2026-05-05/stale-plans-and-orphan-commits.md.

You are composing N parallel returns into one coherent artifact. This is the alchemy step. Stapling returns together is failure; the user explicitly forbade flattening (the grain itself said "resist flattening"). Composition means structural integration, named tensions, resolved disagreements, no si

Distill *how a specific voice thinks and speaks* from a pile of session logs, and persist it as a machine-usable lexicon that downstream agents can write against.

Surfaces the parallel domain dimensions implicit in a dense or minimal prompt. Use when a user prompt is small on the surface but plainly implies multiple independent domains needing different expertise; when explicitly invoked by the coliseum-orchestrator skill as Phase 1; or when the user asks "what dimensions does this prompt encode" or "what axes does this break into." Produces a named dimension set where each dimension is independently executable and not a paraphrase of another.

Foundational layer beneath PDE. Governs eight-strata substrate (ontology, lineage, constellation, gap-map, agent-fleet, production-stack, internal-magnet, external-contribution) plus four operators (selfish-altruistic loop, magnetic membrane, portfolio, reflexive) for any domain engagement. Use when starting a new domain pierce, before product-layer work, when auditing substrate completeness, or when extracting cross-domain patterns via portfolio-pull / portfolio-resonance / reflexive-study / engine-extract modes.

Orchestrate a full directory-graph autopsy across an ORGANVM workspace, then emit migration signals that unite repositories under canonical governance. Thin wrapper over organvm ecosystem + organvm irf + promotion-readiness-checklist. Triggers on auto-autopsy, map all directories, governance migration, ecosystem discovery, or "send the signal to unite" requests.

Enforces ecosystem-wide governance via the 4-tier model and Triple-Reference axiom. Use for governance compliance audits, repository migration planning, triple-reference verification, and dependency-aware migration manifests.

Multi-perspective collaborative inquiry. Decomposes a topic into six cognitive lenses (Scope, Logic, Mythos, Bridge, Meta, Pattern) that operate as a chorus rather than a single voice, then synthesizes meta-patterns no single lens could find alone. MANDATORY TRIGGERS: 'expansive inquiry', 'multi-perspective inquiry', 'explore this from every angle', 'six-lens analysis', 'run a chorus on this'. STRONG TRIGGERS: 'help me think through X deeply', 'mythopoetic AND logical analysis'. Do NOT trigger on simple Q&A or failure-focused inquiries — use premortem for those.

Exports all reports and session transcripts to markdown in a structured archive. Bundles Claude plans, session transcripts, SpecStory history, organized session files, and reports/audits/summaries into a dated export directory with manifest. Triggers on export sessions, archive transcripts, backup reports, or session documentation requests.

Phase 1 of the pentaphase structural-overhaul protocol. Inventories assets, maps current flow, identifies friction, and defines value metrics for any substrate. Use when the user invokes phase 1 of an overhaul, requests a baseline audit, asks to "discover the landscape" of a system, or wants to understand current state before redesigning. Produces phase-1-landscape-report.md.

Every sufficiently long artifact has a passage where its energies converge — the block a reader quotes, a trailer cuts to, an editor pulls. This skill finds that block by scoring rather than vibes, and returns it verbatim with the evidence for why it wins.

Translate internal markdown artifacts (pitch decks, business plans, research) into per-persona client-facing storefront surfaces in the client's domain language with ELI5/TLDR layers. Sits above Product Domain Engine and 8-Strata Domain Ideal-Whole; invokes voice-enforcement against persona-specific voice_constitution. Triggers on requests for client-facing surfaces, persona-tuned translation, ELI5/TLDR projections, or storefront-style client deliverables. Render is scaffold-and-curate (never auto-publish) to protect client trust.

When capabilities are spread across many agent ecosystems (Claude, Codex, Gemini, OpenCode, OpenClaw, a shared `.agents` pool, a skills distribution), the expensive question is no longer "how do I do X" but **"which of my agents already has a skill for X, and how do I load it?"** `portal-router` ans

The meta-system. Every product tied to a domain is an instance of the same operation: formalize informal knowledge into computable structure, then express that structure through four independent rhetorical channels.

Convert loose operator prose into atoms: minimal, classified, machine-readable units of intent that registries can store, pipelines can route, and future sessions can execute without re-interpreting the original wall of text.

Provides structured prompt relay envelopes for receiving and processing feedback from external AI agents. Handles feedback routing, formatting, response generation, and session tracking across agent boundaries. Triggers on cross-agent feedback, external review comments, multi-agent coordination, or feedback relay requests.

1. Read referenced session transcripts/artifacts fresh from disk 2. Verify each claim against on-disk reality (file exists, content matches, commit landed) 3. Mark each as: verified / false-positive / false-negative / partial 4. Produce remediation list with explicit owners 5. Stop at verification â

Strip a system down to what it *is made of*, tier by tier, until it can be compared honestly against everything that already exists — so the verdict "build new" is earned rather than assumed.