Best practices for building an unauthenticated public Q&A chatbot widget. Covers rate limiting, security hardening, cost optimization, semantic caching, observability, UX patterns, chat scroll behavior, and architecture. Tech-agnostic with concrete examples from a production implementation.
Write or revise fast-moving, voice-forward serialized pulp fiction with strong hooks, clear character dynamics, escalating complications, compact chapters, and bingeable momentum. Use when drafting genre fiction, outlining a serial, tightening pacing, improving dialogue, designing chapter cliffhangers, or revising…
Audit or harden an existing application's release readiness, including deploy prerequisites, minimal release gates, rollback, production-shaped smoke checks, and post-deploy verification. Use only when the user explicitly asks for a release-readiness review, deploy or release gates, rollback rehearsal, production…
Draft or substantially revise source-backed nonfiction explicitly requested as a finished writing artifact, such as an article, essay, profile, technical explainer, report, or decision memo. Exclude research-only requests, app playtests, working notes, link collections, and routine factual answers, even when saved as…
Design, implement, audit, or refresh protected username and handle namespaces for public products. Use whenever a product has open signup, mutable handles, profile URLs, impersonation or squatting risk, reserved route names, short usernames, common-word or common-name claims, developer/AI terminology, notable social…
Operate desktop applications and browsers reliably through Computer Use, with proactive state tracking and recovery from common automation failures. Use for any multi-step or consequential UI workflow, especially when app focus moves, multiple monitors or windows are present, accessibility state is empty or stale…
Build or debug applications that specifically use Cloudflare Sandbox SDK (@cloudflare/sandbox) for isolated command execution, code interpretation, files, processes, sessions, or preview URLs. Do not trigger for generic container design, local shell scripting, or unrelated sandbox products.
Design patterns and lessons for building conference schedule pages with React. Covers theme systems, 2D grid overviews, filter UX, sticky layouts, modals, favorites, and data normalization. Use when building or modifying schedule views, grid tables, filter panels, or conference UI components.
Audit or harden a defined application-security attack surface when the user explicitly requests a security audit, vulnerability investigation, or scoped security-hardening pass. Do not trigger for routine authentication changes, role design, scoped permissions, ordinary authorization bugs, dependency updates, generic…
Critically audit or explicitly trim an existing agent skill to its behavioral core, including narrowing over-broad trigger metadata. Use when the user asks to cut, simplify, shorten, de-slop, or reduce the context cost or trigger aggressiveness of a SKILL.md or skill folder. Do not trigger merely because a skill is…
Design or harden Slack bot transport and interaction architecture: signed Events API ingress, fast acknowledgement, idempotency, causal thread context, Block Kit actions, message rendering correctness (Markdown, rich text, links and citations), durable agent execution, multi-workspace operation, or Slack-specific…
Resolve uncertain person or organization identities across aliases, duplicate records, conflicting sources, incomplete identifiers, or competing profile and media candidates. Use when reliable matching requires contextual corroboration, candidate discovery, historical reconciliation, or explicit ambiguity handling. Do…
Summarizes arbitrarily long text (1k-1M words) using recursive map-reduce with any LLM backend. Accepts raw text, markdown, transcripts, articles, codebases, or any plaintext input. Produces one or more output formats: executive summary, section headings with timestamps, YouTube description, Twitter/X posts, title…
Syncs URL query parameters with app navigation, tabs, and filter state so views are bookmarkable and shareable. Use when building or changing frontend routing, filters, tabs, modals, deep links, browser back/forward, or when the user asks to remember navigation in the URL, sync state to query params, or make screens…
Diagnose or improve tests as a system, including flaky or nondeterministic suites, duplicated coverage, expensive fixtures, slow feedback, and test selection. Use only when the user explicitly makes test architecture or suite behavior the primary task. Do not trigger for ordinary implementation work, adding or fixing…
Use when testing the AI Engineer Europe internal Bun schedule preview or public schedule page, including tooltip, modal, CFP metadata, and local preview workflows.
Extracts the most interesting frames from video files for thumbnail compositing. Detects faces, expressions, smiles, and presentation slides. Outputs full frames, face crops, and transparent cutouts. Use when asked to extract thumbnails, find interesting frames, grab screenshots from video, or create thumbnail…
Transcribes audio and video files to text using pluggable ASR backends. Default backend is local whisper/MLX Whisper for ASR. Supports pyannote.audio direct diarization, whisperX, insanely-fast-whisper, faster-whisper, whisper.cpp, OpenAI Whisper API, Groq Whisper API, Deepgram, AssemblyAI, Gemini, and Hugging Face…
Add or repair Cloudflare Turnstile on an existing web form by creating or reusing a widget, embedding it, wiring mandatory server-side Siteverify in the existing backend, and validating the result. Use when the user explicitly asks to add Turnstile, replace reCAPTCHA or hCaptcha with Turnstile, protect a form from…
Use when scraping public Twitter/X timelines or lists through Nitter-style mirrors with axios and Playwright fallback, including anti-bot challenge handling and pagination limits.
Audit a defined Vercel production cost question using billing and usage evidence, then recommend or verify bounded remediations. Use only when the user explicitly asks for a Vercel cost review, unexpected bill or usage spike investigation, Fast Data Transfer analysis, or post-remediation savings verification. Do not…
Turn a recorded talk and transcript into a source-grounded technical article readers can follow alongside the recording, with verified links, useful code, and selective screenshots or explanatory visuals. Use for transcript-to-essay rewrites, illustrated reading versions, and their generation/review pipelines.
Visually inspect a local or deployed site/app in a real browser and report concrete layout, responsive, or design regressions. Use when asked to visually check UI changes, reproduce a screenshot issue, compare a page with a reference, or playtest representative visual states. Do not use for code-only review or…
Prevent layout/transition/reflow thrash in JS-driven CSS animations. Use when building or debugging continuous motion (rotation, orbit, parallax, scroll-linked effects, marquee, carousels) and when symptoms appear like jank, stutter, lag, animation drift, ghosting, "elements collapsing toward center", "going slow"…
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: