54 skills, ready to install.

--- name: super-memory description: > One-command installer for the full Tori memory stack: cognee 1.0.x knowledge graph (pghybrid), LiteLLM proxy with rotation, cognee-api watchdog cron, nightly ingest pipeline, CouchDB 3 single-node document store, and a filesystem markdown vault wired to Obsidian. Install order is encoded in the 11 numbered scripts; the top-level install.sh --detect-first orchestrator runs only what is missing. Use when provisioning a fresh Linux box or replicating the adam

Autonomous TikTok slideshow marketing for any app or product. Onboarding, competitor research, AI image generation, text overlays, Postiz posting, analytics feedback loop, optional RevenueCat conversion attribution.

Run your one-person company on autopilot. Daily morning brief, marketing pipeline, code task delegation, ticket triage, weekly retro with MRR tracking. Battle-tested by WiseChef for 6+ weeks.

Build knowledge graphs from codebases, docs, papers, and images using Graphify. Produces interactive HTML visualization, queryable JSON, and audit report with god nodes, communities, and surprising connections. Use when analyzing unfamiliar repos, understanding architecture, or building structured context before a complex task.

Pull GA4, Google Search Console, Meta Ads, and TikTok Ads data for any client and generate a branded PDF report. White-label-ready. Saves 4h/week per client.

Automated code review with security scanning and best practices enforcement.

AI-powered email drafting with tone control and template library.

High-performance web scraping with built-in rate limiting and proxy rotation.

Interact with X/Twitter via the x-cli terminal client using official X API credentials. Use for posting, reading timelines, searching tweets, liking, retweeting, bookmarks, mentions, and user lookups.

OpenAI's general-purpose speech recognition model. Supports 99 languages, transcription, translation to English, and language identification. Six model sizes from tiny (39M params) to large (1550M params). Use for speech-to-text, podcast transcription, or multilingual audio processing. Best for robust, multilingual ASR.

Iterate on SVG icon/logo geometry using vision-AI feedback in a separate test harness BEFORE touching production. Covers the test-harness-first pattern, vision-inconsistency mitigation (same path can score 9/10 then 4/10 across calls), the "vision sees lobes wrong" calibration step, and the rule that human-eyeball-on-real-browser is the final gate for animated/aesthetic work. Use when designing iconography (chef hats, logos, marks), writing parametric SVG paths, or any visual iteration where you need objective feedback before deploying.

AI-optimized web search via Tavily API. Returns clean, relevant content designed for AI agent consumption. Supports deep search and news.

Summarize any content via the summarize CLI — URLs, PDFs, images, audio files, YouTube videos. Multi-provider LLM support.

State-of-the-art text-to-image generation with Stable Diffusion models via HuggingFace Diffusers. Use when generating images from text prompts, performing image-to-image translation, inpainting, or building custom diffusion pipelines.

Generate visual diagrams and hero images from code repositories using GitNexus + D2 + Playwright pipeline. Great for READMEs, blog posts, and architecture overviews.

Generate structured PR descriptions from git diffs using Conventional Commits format. Analyzes branch diff, commits, and status to produce ready-to-paste PR descriptions.

Edit PDFs with natural-language instructions using the nano-pdf CLI. Modify text, fix typos, update titles, and make content changes to specific pages without manual editing.

Generate or edit images via Google Gemini 3 Pro. Create images from text, edit existing images, or compose multiple images. Cost-effective alternative to DALL-E/gpt-image.

|

Guide for building production-quality MCP servers in Python (FastMCP/Pydantic) or TypeScript (MCP SDK/Zod). 4-phase process — Research, Implement, Review, Evaluate.

Search 100k+ indexed AI skills, tools, agents, and apps by natural language task description. "Google for AI capabilities.

Production pipeline for HTML-based video rendering using HeyGen's HyperFrames framework. HTML plus CSS plus GSAP rendered to deterministic MP4 via headless Chrome and FFmpeg. Agent-native with no React and no bundler. Use when generating marketing reels, educational carousels-as-video, animated explainers, or any composition where you would reach for Remotion. HyperFrames is Apache-2.0 licensed and frame-accurate for GSAP timelines.

Production pipeline for vertical animated invitation cards (wedding, save-the-date, birthday, event) using HyperFrames. 1080x1920 H.264 MP4 under 4MB that plays inline on WhatsApp, Messenger, Instagram Stories, and email. Covers the 6-scene template (hero, names + circular photo cameo, date, venue/ceremony, reception, closing), music-bed preprocessing with ffmpeg, and the bilingual-fork workflow for shipping the same invitation in multiple languages. Read BEFORE authoring any personal invitation — the pattern is different from the marketing/explainer use case that `hyperframes-video` documents.

Search the web and X/Twitter using xAI's Grok API with real-time access, citations, and image understanding. Requires XAI_API_KEY.

Analyze codebases using GitNexus — a knowledge graph engine that indexes every dependency, call chain, cluster, and execution flow. Use for bug hunting, architecture reviews, and understanding unfamiliar codebases.

Frontend development guidelines for React/Next.js/Tailwind — mobile-first, typography, color theory (70-20-10), accessibility, performance targets, memorable design elements.

Local, offline speech-to-text using CTranslate2 reimplementation of OpenAI Whisper. 4-6x faster, 99+ languages, word-level timestamps, VAD support.

Advanced ElevenLabs TTS — text-to-speech, voice listing/filtering, credits management. Script-based, works standalone.

Model software around the business domain using bounded contexts, aggregates, and ubiquitous language. Use when the user mentions "domain modeling", "bounded context", "aggregate root", "ubiquitous language", or "anti-corruption layer". Covers entities vs value objects, domain events, and context mapping strategies. For architecture layers, see clean-architecture. For complexity, see software-design-philosophy.

Adversarial code review — scrutinizes code for security holes, lazy patterns, race conditions, edge cases, and bad practices. Structured output with Blocking/Required/Suggestions severity tiers.

Generate project ideas through creative constraints. Use when the user says 'I want to build something', 'give me a project idea', 'I'm bored', 'what should I make', 'inspire me', or any variant of 'I have tools but no direction'. Works for code, art, hardware, writing, tools, and anything that can be made.

Guidelines for performing thorough code reviews with security and quality focus

Write readable, maintainable code through disciplined naming, small functions, and clean error handling. Use when the user mentions "code review", "naming conventions", "function too long", "code smells", or "readable code". Covers SRP, comment discipline, formatting, and unit testing. For refactoring techniques, see refactoring-patterns. For architecture, see clean-architecture.

Structure software around the Dependency Rule: source code dependencies point inward from frameworks to use cases to entities. Use when the user mentions "architecture layers", "dependency rule", "ports and adapters", "hexagonal architecture", or "use case boundary". Covers component principles, boundaries, and SOLID. For code quality, see clean-code. For domain modeling, see domain-driven-design.

Use before any creative work — designing features, building components, adding functionality, or modifying behavior. Explores user intent, requirements, and design before implementation. Especially valuable when scope is ambiguous, multiple approaches exist, or the work touches unfamiliar codebases.

Lighthouse + content audit + competitor diff. Generates branded PDF or markdown report.

Generate images from text prompts using multiple AI models.

Visual pipeline builder for ETL workflows with 50+ connectors.

Pre-flight checklist before declaring any deployment "live" — DNS resolves to your tunnel, origin code is current commit, alembic head matches, smoke tests on PUBLIC URL not localhost.

Diagnose and fix Stripe webhook handlers crashing with AttributeError get or KeyError get after a Stripe Python SDK 15.x upgrade. SDK 15.x StripeObject (including Event) no longer supports dict-style .get() / [] mapping access cleanly — every handler that does event.get(type) or event[data][object] dies on the first webhook delivery. Symptom from the user side is "payment thinking… then error message". Fix is to convert the verified Event to a plain dict immediately after stripe.Webhook.construct_event so all downstream handlers work unchanged. Validated 2026-04-29 on recipes.wisechef.ai (4× consecutive 500s → 200 OK in one patch + restart).

Rotate live Stripe prices on a running production product (e.g. Scan free / Rescue $7 / Ward $29/mo) without breaking checkout. Covers the archive-not-delete reality, the dual env-var coupling (modern + legacy alias), the multi-file frontend copy hunt, the system-vs-user systemd detection, and the JS-bundle-not-just-HTML smoke test. Use when Adam asks "set the pricing to X/Y/Z" on a Stripe-wired product and the change must go live in one session. Validated 2026-04-28 on rescue.wisechef.ai (8min end-to-end).

A comprehensive workflow for researching a startup idea and generating a full startup kit including market analysis, competitor mapping, product specs, and a high-fidelity landing page.

Diagnose a Railway-hosted service that "looks dead" before declaring an outage. Railway serverless mode returns HTTP 404 "Application not found" during the ~3-second cold-start wake window after idle, which is easily and repeatedly misdiagnosed as a deleted/crashed deployment by autonomous agents. Use BEFORE filing infrastructure tickets, opening blocked Paperclip issues, or planning a redeploy. Validated 2026-04-24 on AgentPact (api/mcp/web all "dead" → all healthy after wake).

Define product positioning by mapping competitive alternatives, unique attributes, and best-fit customers to the right market category. Use when the user mentions "positioning", "competitive alternatives", "how to position", "market category", or "why customers don''t get it". Covers positioning canvas and team workshops. For customer jobs analysis, see jobs-to-be-done. For go-to-market, see crossing-the-chasm.

Talk to customers without leading them using Mom Test rules: discuss their life not your idea, ask about specifics in the past, and talk less. Use when the user mentions "customer interviews", "validate my idea", "users say they want it but don''t buy", "leading questions", or "The Mom Test". Covers commitment and advancement, avoiding compliments, and extracting signal from noise. For product-market fit, see jobs-to-be-done. For rapid prototype testing, see design-sprint.

Design MVPs, validated learning experiments, and pivot-or-persevere decisions using Build-Measure-Learn. Use when the user mentions "MVP scope", "validated learning", "pivot or persevere", "vanity metrics", or "test assumptions". Covers innovation accounting and actionable metrics. For 5-day prototype testing, see design-sprint. For customer motivation analysis, see jobs-to-be-done.

Create irresistible offers using the Value Equation, bonus stacking, risk-reversing guarantees, and ethical scarcity. Use when the user mentions "pricing strategy", "irresistible offer", "bonuses and guarantees", "value-to-price ratio", or "offer naming". Covers the MAGIC naming formula and starving-crowd targeting. For product positioning, see obviously-awesome. For outbound sales, see predictable-revenue.

Read and edit DNS records on Hostinger-registered domains via the Hostinger Developer API. Use when a domain's nameservers are ns1/ns2.dns-parking.com (Hostinger default), when registrar lookup via RDAP returns "HOSTINGER operations, UAB", or whenever you need to script DNS changes without logging into hpanel. Validated 2026-04-24 on agentpact.xyz (CNAME swap to update Railway custom-domain target).

Combined customer discovery + competitive intelligence analysis for a product or startup niche. Produces: niche pain profiles, early adopter personas, discovery questions, validation criteria, vitamin/painkiller verdict, cobbled-solution audit, competitor map with real pricing, indirect competitor landscape, moat assessment. Use when user says 'validate the problem', 'map the competition', 'customer discovery for X', or 'competitive analysis for Y'.

Diagnose and fix Caddy serving a static multi-page site (Astro, Next static-export, Eleventy, Hugo) where every route returns the homepage. Covers the SPA-fallback `try_files` trap, the silent-no-op port conflict (manual caddy process holding the port while systemd thinks it's down), and the caddy-user-cannot-traverse-home-directory permission trap. Use when "links don't work," "all routes return same content," or "caddy says active but pages 404 / serve wrong content.

Build and roll out a stack-agnostic brand meta-repo across multiple consumer products. Use when unifying design/brand/assets across different web stacks (Astro + React + Vite + others) without forcing them onto the same framework. Establishes a single source of truth for tokens, logos, Tailwind preset, and ships an auto-propagation workflow that opens bump PRs in every consumer repo on tag.

Spin up a per-client Streamlit dashboard with live GA4/GSC/Meta/TikTok metrics. Branded, password-protected.

Personalized cold email generator with MX deliverability check. Optional safe SMTP send.

Discovery transcript → branded SOW PDF with scope, pricing tiers, case studies. Powered by Anthropic Claude.