cuga-apps

Sales-dev scout for finding local businesses that would benefit from a conversational AI agent. Given a place (and optional category or pitch focus), produces a ranked board of independent prospects with deep-dive evidence and tailored cold emails for the top 3. Triggers on "find leads in [place]", "scout [city]", "[category] in [place] who need [capability]".

Generate human-readable API documentation from an OpenAPI / Swagger spec on disk. Lists endpoints, expands schema $refs, and writes per-endpoint sections with realistic curl examples and response shapes. Use when the user asks to "document this API", "generate docs for <spec.json|spec.yaml>", or paste an OpenAPI spec path.

Generate Mermaid.js architecture diagrams from natural-language system descriptions. Returns a fenced ```mermaid block ready for any markdown renderer, plus a brief components explanation. Use when the user asks for a diagram, flowchart, sequence diagram, ER diagram, or wants to "draw" a system.

Browse a Box folder and answer questions over its documents (PDF, DOCX, PPTX, XLSX, TXT, MD, CSV) with file-cited answers. Use when the user references "Box", a Box folder ID, or asks "find/show me <file> in my Box".

Produce a structured research brief on a question using a stated tool-call budget. You decide the decomposition, allocate budget across sub-topics, and synthesise sourced findings. Use when the user asks for a brief, primer, or literature snapshot with an explicit budget on lookups.

Assemble a one-screen city briefing — weather, today's news, encyclopedia background, optional attractions, optional crypto spotlight — for any city the user names. Use when the user asks for a "briefing", "what's happening in", or "tell me about <city> today".

Review a code snippet (or whole file) and return structured, actionable feedback covering bugs, security flaws, performance, style, and architectural insights. Use when the user pastes code and asks for a review, audit, critique, or "look this over".

Extract and summarise the contents of a local document the user supplies by file path (.txt, .md, .pdf, .docx, .pptx, .xlsx, .csv). Use when the user uploads, drops, or names a path to a file and wants a TL;DR with key points and action items.

Recommend real IBM Cloud services for a described use case, explain how they connect, and provide ibmcloud CLI commands. Use when the user asks "which IBM service for…", "AWS X equivalent on IBM Cloud", or wants to design an IBM Cloud architecture.

Answer IBM Cloud / IBM product questions by searching real IBM documentation and synthesising sourced answers. Use when the user asks "how do I…" or "what does <IBM service> do" with an IBM Cloud / Watson / Power / Z context.

Track and digest IBM Cloud release notes / "What's New" announcements for one or more IBM services. Use when the user asks "what's new in <IBM service>", "recent updates to <X>", or wants a release-notes digest.

Recommend 5–8 films for a user based on their stated taste profile (liked films, genres, directors, actors, mood, dislikes). Verifies movie facts via Wikipedia. Use when the user asks "what should I watch", "recommend films", or describes their taste and wants suggestions.

Fetch RSS / Atom feeds the user supplies, optionally filter by keywords, and produce a digest of recent items with links. Use when the user names feed URLs and asks for a "digest", "what's new in my feeds", or wants keyword-filtered headlines.

Discover and summarise academic research papers using arXiv and Semantic Scholar. Use whenever a user asks for papers on a topic, pastes an arXiv ID/URL, or wants to know what a paper builds on.

Suggest 3–5 cookable recipes for tonight from a user-supplied pantry, respecting diet (vegetarian, vegan, keto, etc.) and allergies. Use when the user names ingredients they have and asks "what should I cook" / "recipe ideas" / "dinner suggestions".

Look up current crypto and stock prices with 24h change. Crypto is keyless via CoinGecko; stocks need an Alpha Vantage key. Use when the user asks for a price, "how much is BTC / AAPL / TSLA", a watchlist status, or a quick comparison.

Generate a structured multi-day travel itinerary for any destination by combining Wikipedia background, real weather, geocoding, attractions search, and live web search. Use when the user asks to "plan a trip", "build an itinerary", or "X-day itinerary for <place>".

Design a multi-day travel itinerary with a goal-shaped planner — you decide the decomposition, themes, and tool order. Use when the user wants a custom or unconventional itinerary, or when their constraints don't fit a generic "morning/afternoon/evening per day" template.

Run a one-shot web research pass — multiple targeted queries, synthesise findings, cite sources. Use when the user asks "research X", "what's the current state of Y", or "find me info on Z" and wants a structured report (not just a single page summary).

Fetch any webpage and produce a structured summary — title, overview, key topics, important facts, and a one-line takeaway. Use whenever a user pastes a URL and asks "summarize", "what's on this page", or "what does this say".

Deep Wikipedia research — go beyond the keyword search by reading articles section-by-section, following related links, and synthesising a structured report with citations. Use when the user wants a deep dive, primer, or thorough background on a topic.

Research a topic via YouTube — find relevant videos, fetch their transcripts, and synthesise findings with timestamped citations. Or paste YouTube URLs directly for instant transcript-based summaries.

Discover, filter, and evaluate hiking trails near any location using OpenStreetMap. Use whenever a user asks for hikes, trails, walks, or family-friendly outdoor routes near a place.

Synthesize the final ranked lead board and a tailored cold email per deep-dived lead. Pure synthesis — no external lookups. Use last, after scout/site_auditor/voice_of_customer/person_finder/stack_scanner/revenue_estimator have produced their structured outputs.

Mine review-site snippets and complaint posts for verbatim friction quotes about a specific business. Use after scout has named a candidate, in the deep-dive phase, to ground pitches in real customer pain rather than abstractions.

Find a likely decision-maker (owner / manager / GM) for a business and propose a best-guess direct email with a confidence rating. Use after scout has named a candidate, when the cold outreach needs a real person not a generic info@.

Estimate the annual-revenue band of a business from public size signals (review counts, employee mentions, hours×days density, multi-location flags). Output is a coarse band, not a number, with explicit "estimated, not measured" disclaimer. Use to rank leads by recoverable revenue.

Resolve a place name to coordinates and surface candidate local businesses by category from OpenStreetMap. Use as the very first specialist on any new lead-hunt request.

Fetch a business website and classify it on capability gaps (no online ordering, phone-first, no chat) and freshness flaws (no HTTPS, no mobile, stale copyright, dated tech stack). Use after scout has surfaced a candidate with a website URL.

Fingerprint third-party tools embedded on a business's website (OpenTable, Calendly, Toast, Square, Resy, Zocdoc, etc.). Tells the pitch writer whether to argue "displace X" or "you have nothing — start here". Use after scout has named a candidate with a website URL.