AI Agent Workflow Collection

Generate tiered sentence frames for academic tasks that scaffold language production across proficiency levels. Use when EAL students need structured language support for classroom discourse.

Use when the question is whether users can operate a UI — keyboard, switch device, voice, screen-reader gestures — not just by mouse or touch. Trigger when designing keyboard navigation, focus management for modals/menus/popovers, drag-and-drop with alternatives, motion-heavy interactions, or any flow that asks the user to do something. Covers WCAG Principle 2 (Operable). Sub-aspect of accessibility; read that first if you haven''t already.

Use when the question is whether users can perceive the content — see it, hear it, or feel it — regardless of impairment. Trigger when designing color systems, picking text sizes, building image-heavy surfaces, working with video/audio, designing for low-vision users, or running an accessibility audit. Covers WCAG Principle 1 (Perceivable). Sub-aspect of accessibility; read that first if you haven''t already.

Use when the question is whether assistive technologies — screen readers, voice control, switch devices, refreshable braille displays — can reliably interpret your UI. Trigger when picking between semantic HTML and a custom-built component, when reaching for ARIA, when designing live regions for dynamic content (toasts, validation, search results), or when reviewing a UI built largely from elements. Covers WCAG Principle 4 (Robust). Sub-aspect of accessibility; read that first if you haven''t already.

Use when the question is whether users can understand the UI — predict its behavior, recover from errors, comprehend its labels and copy. Trigger when writing form labels and error messages, when designing flows with surprising state changes, when picking microcopy for the destructive moments, or when reviewing a UI that "works correctly but confuses people." Covers WCAG Principle 3 (Understandable). Sub-aspect of accessibility; read that first if you haven''t already.

Generate a cascading hint sequence for a problem type, revealing progressively without giving answers. Use when designing tutoring dialogues or scaffolded worksheets.

Use when the design''s perceived quality affects whether users will tolerate friction, give the product a chance, or judge it as well-made — which is most consumer-facing surfaces, marketing, onboarding, first-run, and any moment where an unfamiliar user is forming a quick judgment. Trigger when designing landing pages, sign-up flows, first-time UX, app store screenshots, demo videos, sales decks, or any surface where the user''s first impression matters and friction is unavoidable. Also trigger when defending why "polish" matters on a working product, or when the user asks "should we ship the spartan version or invest in design?

Use when a design task involves the felt quality of an interface — beauty, polish, brand voice, persuasion, emotional response, marketing surfaces, hero sections, onboarding tone, empty/celebration states, illustrations, mascots, avatars. Trigger when the user mentions tone, polish, brand, "feels cold/cheap/premium," personality, voice, mascot, hero, "make it more delightful," or asks how a screen should make the user feel. Framework-agnostic. Routes the model to the right aesthetics principle in this plugin.

Use when the question is whether the user can tell what an element does — whether it looks clickable, draggable, expandable, editable, or interactive at all. Trigger when designing buttons, links, custom controls, drag handles, draggable cards, editable inline fields, or any interactive element. Trigger on reviews when users hesitate, miss controls, or click on things that aren''t interactive. Affordance is one of the most decisive principles in interaction design — Norman placed it at the center of The Design of Everyday Things — and it''s one of the most violated by modern flat-design conventions that strip the visual cues users rely on.

Use when designing or fixing controls that mislead users — elements that look interactive but aren''t (false affordance), or controls that need to clearly communicate "you can''t do that here" (anti-affordance, like disabled states). Trigger when reviewing UIs where users tap inert elements expecting response, when designing disabled states, when picking treatments for read-only fields, or when a "decorative" element is being mistaken for a button. Sub-aspect of affordance; read that first.

Use when the design has interactive capability but users aren''t seeing it — when the affordance is technically present but the signifier (the visible cue that tells users "you can do this") is missing or weak. Trigger when reviewing flat-design UIs where everything looks the same, when designing custom controls, when stakeholders ask "should this be more obvious?", or when picking signifier strength for primary vs. secondary actions. Sub-aspect of affordance; read that first.

Map control, influence, and concern after systems analysis. Use when students need wise agency without being made responsible for everything.

Generate scaffolds that gradually increase student choice, voice, and ownership within a learning task. Use when students depend heavily on teacher direction and need to develop autonomy.

Use when the user wants to benchmark or compare AI agents (Claude Code, Codex, OpenCode) on a refactoring, perf, or code-change task in the current repo. Use when user says compare agents, benchmark Claude vs Codex, agent eval, measure agent, AI agent comparison, agent trial, /ai-agent-bench.

After any AI-generated explanation, require the learner to identify one place it could be wrong, one thing to check, and one source to consult. Builds epistemic vigilance — treats AI output as a claim to evaluate, not truth to absorb.

Design a Funhouse Mirror activity where students use their own domain expertise to detect AI distortions, omissions, and overconfidence. Use when students know a subject well enough to evaluate AI claims about it.

Audit and redesign AI-generated feedback for pedagogical quality, timing, and learning impact. Use when building or reviewing automated feedback in digital learning tools.

Design a fact-checking protocol for AI-generated text, extending SIFT with AI-specific adaptations for hallucination detection. Use when students need to verify AI claims and citations.

Map which elements of an assignment benefit from AI assistance vs. which AI use undermines. Use when redesigning tasks for AI-age classrooms or setting defensible AI use policies for specific assignments.

Design a structured protocol for auditing AI-generated text against Ennis's six CT standards. Use when students need to critically evaluate AI output in any subject.

Design a multi-round questioning sequence for interrogating AI chatbot answers, tracking how responses shift and distinguishing genuine updates from sycophantic capitulation. Use when teaching students to probe AI critically.

Use when the user asks Codex to inspect Aient telemetry, search logs, explain traces, create/list Aient API keys, or use the Aient MCP server.

HTTP API-based documentation server (always running). Questions about project architecture, conventions, decisions, code structure, tech debt, env config, progress, or doc health. Also: init project, audit docs, lint, validate, promote note, rebuild index, search vaults.

Codex-native supplier sourcing and outreach workflow for Alibaba using LaunchFast research. Use when the user wants supplier shortlists, outreach messages, reply triage, or negotiation support.