Tools

Complete command library for Ada, a two-mode software design document consultant. Covers all commands, phases, modes, and examples for building rigorous SDDs from problem definition to ship-ready spec.

Bacon is a five-stage academic writing coach that converts raw project descriptions into peer-reviewable scientific papers. Built for researchers, graduate students, and academic writers who need structured guidance — not confident-sounding prose over invisible gaps.

Complete command library for Bookie, a two-mode Materials Science and Engineering textbook author. Covers all commands, authoring standards, phase gates, and pushback behavior for writing publication-ready chapter drafts.

Cajal runs eight-phase forensic audits on scientific, medical, and AI-generated images. Every finding is severity-rated, every Critical or Major issue paired with a reparative prompt. Two modes: silent execution or active expert interrogation.

CAZE is a dual-mode consulting tool that generates fictional case interview scenarios and rigorous real-company strategic diagnostics. No fuzzy thinking, no invented data — every claim sourced or labeled, every contradiction flagged before a word of analysis is written.

Courses turns any source content into course-ready artifacts: slide decks, video storyboards, learning outcomes, assessments, and visual production recipes. Applies backwards design, Bloom

CRITIQ is a two-mode peer review and manuscript development tool covering the full paper lifecycle: brainstorming a testable research question, drafting IMRaD sections, running eight-dimension peer review, and preparing point-by-point reviewer responses. Built for researchers at every level.

Complete command library for Eddy the Editor, a two-mode Substack editorial mentor for student writers. Covers all commands, the seven-stage review framework, phase gates, pushback behavior, and the 60-Minute Surge Protocol.

Complete command library for Eddy the Storyboarder, a two-mode CapCut explainer video architect. Covers all commands, scene architecture, intake protocol, pushback behavior, and production notes for building production-ready storyboards.

Complete command reference for Figure Architect, a two-mode scientific visualization tool. Scans text for high-assertion zones, generates Hero Images, and produces publication-quality prompt sets for BioRender, Illustrae, and Midjourney. For researchers, science writers, and technical communicators.

Iris is a command-based editorial art direction engine for long-form magazine hero images. She translates a journalism article

Madison builds ten-slide venture pitches for AI-driven tools — running jargon audits, enforcing timing discipline at 130 WPM, and stripping AI scaffolding before anything reaches an investor, donor, or grader. Two modes: silent execution or active expert pressure.

Mega transforms end-of-chapter exercises for quantitative courses into LLM-integrated problem sets that teach students which prompting strategy to use and why. Three strategies, deliberate distribution, a pushback layer that catches the most common exercise design error.

Méliès transforms songs, stories, and images into phone-footage-style prompt sequences for AI image and video generation. Named after Georges Méliès, who made impossible things look documented. Two modes: silent execution or active directorial guidance.

Complete command library for the NEU Writing Coach, a two-mode Socratic writing instructor for Northeastern University masters and MBA students. Covers ten college commands, direct diagnosis protocol, phase gates, pushback behavior, and genre-specific failure patterns.

PROOF audits what a non-profit brand actually establishes — not what it claims, intends, or aspires to. Five commands, two modes, every finding labeled, every recommendation traceable to a specific observation. Built for 501(c)(3) organizations that need to know what their brand actually proves.

A two-mode profile intake interviewer that gathers raw material for long-form stories through conversation, not extraction. Runs in Rose mode (probing, synthesizing, unhurried) or Terkel mode (short questions, long silences, subject

Sherpa converts experiential placements into structured learning evidence. Weekly progress reports, MVAL journal entries, compliance audits, and capstone narratives for co-op, internship, clinical rotation, study abroad, and practicum. Part of the AI Sherpa infrastructure.

A two-mode AI syllabus architect for faculty and instructional designers. Covers the full pipeline from course concept through accessibility compliance — intake, outcomes, arc design, alignment audits, policy writing, and finalization diagnostics. Built for syllabi that teach, not intimidate.

Complete command library for SOCRIT, a two-mode Socratic prompt engineer. Covers all commands, the eight-section evaluation framework, Paul-Elder element coverage, linguistic transformation rules, learner state classification, domain heuristics, and psychological safety audit.

Subby is a two-mode Substack writing assistant covering nineteen format-specific commands — from 300-word Notes to 3,000-word serial installments, subject lines to manifesto statements. It asks before it writes, pushes back on weak briefs, and will not produce a take without a specific claim.

A two-mode AI textbook architecture consultant for authors, editors, and curriculum designers. Applies backward design, Bloom

Complete command library for WAYPOINT, a two-mode experiential learning report evaluator. Covers all commands, compliance framework, developmental framework, five domain-specific tracks, MVAL protocol, output registers, and pushback behavior for co-op, study abroad, clinical, apprenticeship, and corporate early career programs.

Zelda is a two-mode GDD consultant with 20+ years shipping titles across AAA, indie, and mobile. She puts a senior game designer in the room — asking before acting, pushing back on vague concepts before they become production debt, and holding phase gates between vision, systems, world, and scope.