Run an editorial review of the Cyberneutics repository (essays and supporting material) against the seven-dimension audience-experience rubric. Produces rubric scores with evidence, per-essay notes, cross-cutting issues, and a prioritized remediation plan. Does not rewrite content. Use when the user types '/editorial-review' or asks for an editorial review of the repo.
Write diary entries for wild/diary/ — the repo's rawest exploratory record. Entries capture ideas from conversations in connected prose with cross-references, open questions, and public-facing privacy filters. Use when the user types '/diary' or asks to write a diary entry, or when a conversation produces exploratory ideas that should be recorded.
Generate a session handoff document for successor agents. Use when the user types '/handoff' or asks to create a handoff. Reads the previous handoff (if any) for continuity, generates a new handoff with today's date, saves it to agent/, and archives the previous one to agent/archive/.
Convert resource equations into Mermaid string diagrams. Use when the user has typed resource equations (A × B → C style) and wants a visual diagram, or when the user describes a pipeline of operations on typed artifacts and wants both the equations and the diagram. Also use when editing or extending an existing set of resource equations.
Run an adversarial committee deliberation using the roster defined in agent/roster.md to explore decision spaces, surface assumptions, and map trade-offs. Every run writes a deliberation record to <situation-dir>/deliberations/ (00–04 files). Use when the user types '/committee [topic]' or asks for a committee deliberation.
Run an independent review of a committee deliberation transcript. Evaluates against the five core rubrics, cites specific transcript passages, and produces actionable feedback. Use '/review' after a committee run or with a pasted transcript; use '/review --situation <path>' to review that situation's deliberation and write transcript_review to 04-evaluation-1.md (or 06-evaluation-2.md, 08-evaluation-3.md when the feedback loop has run).
Run the composed fan→funnel pipeline N times on the same situation to map the decision landscape. Produces a variance report (what's stable vs. variable across runs) and a decision landscape map (basins, ridges, load-bearing assumptions). Writes to <situation-dir>/probes/ with self-contained run-0N/ subdirectories. Use when the user types '/probe [situation]' or asks to probe a decision space.
Run divergent scenario generation (the fan operation) using the roster defined in agent/scenario-roster.md. Takes a situation description and produces a set of distinct narrative scenarios exploring different possible futures. Every run writes a scenario record to <situation-dir>/scenarios/ (00–03 files). Use when the user types '/scenarios [situation]' or asks for scenario generation.