agent-sync-public

[Cali] Generate and maintain project AGENTS.md files using semantic analysis. Triggers when: a project has no AGENTS.md, user asks to 'create AGENTS.md', 'generate agents md', 'setup agents md', or when starting work on a project that lacks one. Also triggers when user asks to 'check if AGENTS.md is stale', 'update agents md', or when evolving an existing AGENTS.md. Covers: initial generation from codebase analysis, sem-based staleness detection, pre-commit hooks, and CI guards. Always offer hook installation when invoked for any reason.

[Cali] Validate project AGENTS.md files against best practices and the canonical template. Use when: user says 'validate agents md', 'check agents md quality', 'audit my agents md', or after creating/updating AGENTS.md. Checks: structure, size, content quality, template compliance, and provides fix recommendations. Can offer automatic corrections via ask tool.

[Cali] Tool priority rules for code navigation: Cymbal-first, FFF fallback, batch guidance for token efficiency, and semantic diff workflow. Activates on any file-task within a Git repository to guide agent tool selection.

[Cali] Use this skill whenever the user wants to analyze a codebase and generate a detailed product spec document. Triggers for: "analyze this codebase", "generate a spec", "document this project", "reverse engineer this app", "what does this system do", "create product documentation", "map out the features", "extract product rules from code", "what does this codebase do", "generate product spec from code", "document this app", "spec out this repo", "read this code and tell me what it does", "create a product spec", "reverse engineer the product", "what features does this app have", "document the flows", "extract the prompts from this code", or any request to extract product knowledge from source code. Supports any input method: file uploads, pasted code, filesystem paths, or zip archives. Always use this skill when the user shares code AND wants to understand what the product does — even if they don't explicitly say "spec" or "documentation".

[Cali] Go web stack: Datastar, Templ, DaisyUI, NATS, PocketBase/SQLite, GoAI LLM, Zenflow agents. Scaffold, real-time, hypermedia, auth, DB, embeddings, voice AI, AI agents, durable workflows, queues. Queues use NATS JetStream or goqite. Workflows use go-workflows, turbine, ebind, dagnats, or Hatchet per decision tree below.

Use this skill for any Go (Golang) backend task — writing services, APIs, CLI tools, concurrent code, or reviewing/refactoring existing Go code. Triggers on: any .go file, mention of goroutines, channels, mutexes, Go modules, or requests to build backend systems in Go. Also triggers when setting up linters, running tests, managing Go dependencies, or running the app locally during development.

Builds reactive web applications with StarHTML, a Python-first framework over Datastar. Use when writing StarHTML components, signals, event handlers, reactive attributes, conditional helpers, CSS classes, computed signals, HTTP actions, SSE endpoints, or plugins (persist, scroll, resize, drag, canvas, motion, markdown, katex, mermaid, split, nodegraph). For UI components, use StarUI (shadcn/ui for Python). After generating any StarHTML file, run `starhtml-check <file.py>` to validate.

Adds real-time broadcasting, multi-user collaboration, and database persistence to StarHTML applications. Use when implementing chat systems, collaborative editors, presence tracking, typing indicators, or any feature requiring synchronized state across clients. Supports core StarStream (broadcasting), PocketBase (persistence), and Loro (CRDT) plugins.

[Cali] Cross-cutting debugging rules: framework JS API verification and binary freshness checks. Activates when debugging reactive framework interactions (Datastar, Alpine, Vue, Svelte, Solid) or diagnosing code changes that don't take effect after restart.

Reads past DegustIA client report files (HTML or Markdown), extracts every recommended tool and client context, deduplicates tools across clients, then searches Reddit, X, Hacker News, Exa Web, and the wider web for meaningful updates about those tools or related solutions. Produces a prioritized brief of what changed and which specific clients might benefit. Integrates with the last30days skill for social/community signal detection.

Guia a coleta de depoimentos de clientes do Sommelier de IA no momento certo do processo, usando a abordagem de Hormozi: pedir depois da primeira evidência de resultado, nunca na entrega. Gera depoimentos mais autênticos e reduz a sensação de que o cliente está sendo "solicitado".

Gera os relatórios do "Sommelier de IA" (Cali · calionauta.github.io/sommelier-ia) a partir de um transcript de entrevista com cliente. Use SEMPRE que houver um transcript/gravação de conversa com um cliente e a intenção for produzir o relatório da Sutil de 20 min ou do Encorpado pago de 45 min — mesmo que o usuário não diga "skill" nem "relatório" (ex.: "analisa essa entrevista", "monta o report do cliente X", "faz o diagnóstico de IA desse transcript", "resume essa call e sugere ferramentas"). Produz um HTML autocontido com a marca Sommelier (creme/terracota, DM Serif Display + Inter), pronto para virar PDF, em pt-br, priorizando ferramentas simples que o cliente já tem, com cálculo de ROI realista em FAIXAS e um upsell honesto no fim. Os preços dos serviços vêm de `references/config.yaml` — nunca hardcode. NÃO use para escrever posts, e-mails soltos ou para diagnósticos que não venham de um transcript de entrevista.

Extrai métricas estruturadas, cálculos e estimativas de transcripts de entrevistas com clientes do Sommelier de IA. Produz um JSON com dores, frequências, tempo gasto, pessoas envolvidas, economia potencial, ROI e recomendações financeiras. Projetado para alimentar o cali-degustia-diagnostico ou integrar com dashboards/planilhas.

Evaluates existing landing page copy (or a description/brief of a landing page not yet written) and/or generates a new landing page structure with full copy, based on copywriting frameworks mapped to 5 product/service/brand categories, with a focus on human tone and anti-marketing-bullshit writing (avoiding generic-AI clichés like 'elevate', 'unlock', 'seamless', 'journey', empty promises). Use this skill whenever the user mentions a landing page, sales page, opt-in page, conversion copy, or a product/service page structure, or asks to review, critique, improve, or create copy for a page meant to convert — even if they don't use the words 'landing page' explicitly (e.g. 'my course page', 'my SaaS homepage copy', 'copy to attract mentorship clients').

[Cali] Deploy to a private server via Tailscale OIDC with ephemeral credentials. Use when: setting up CI/CD deploy pipeline, configuring Tailscale SSH + OpenSSH for GitHub Actions, creating deploy users with restricted access, writing deploy.yml workflows, or troubleshooting deploy authentication failures. Triggers: "deploy to server", "tailscale deploy", "deploy pipeline", "CI/CD deploy", "deploy workflow", "deploy setup", "deploy key", "deploy user".

[Cali] - INTERACTIVE SKILL: Discover servers from ~/.ssh/config, auto-detect hosts (filtering out non-server entries like github.com), prompt user to pick one via question tool, then SSH into the chosen server and render a real-time ASCII dashboard with Docker containers, images, volumes, routes, cron, orphaned resources, and cleanup suggestions. REQUIRES question tool, SSH config parsing, and shell execution.

Create GitHub releases following project conventions. Triggers when: user says 'release', 'create release', 'push release', 'deploy to main', 'merge to main', user merges a PR to main, or when git push to main is detected. Also triggers on mentions of: gh release, semver, version bump, changelog, release-please. Covers: config-driven (read .release.yml and execute) and fallback (gh CLI) release flows, versioning rules, tag management, and the mandatory release-on-merge convention.

Run supply chain security scans before installing packages or before releases. Triggers when: user installs a package (npm, pip, go get, brew), user asks to 'scan dependencies', 'check vulnerabilities', 'supply chain', 'security audit', 'run trivy', 'run socket', or before any release/deployment. Also triggers on mentions of: socket.dev, trivy, OSV-scanner, dotenvx, CVE, dependency audit. Covers all four tools with concrete commands.

[Cali] Server security audit and hardening for private servers behind Tailscale. Use when: auditing server security, hardening SSH/firewall/Docker, checking for vulnerabilities, setting up fail2ban, reviewing port exposure, responding to security alerts, or storing API tokens/secrets on the server. Covers 7 layers: CloudFlare, UFW, Tailscale, SSH, Docker, Application, Secrets. Triggers: "server security", "security audit", "harden server", "SSH hardening", "firewall rules", "UFW config", "fail2ban", "port security", "Docker security", "vulnerability check", "security review", "API token", "secrets storage", "~/.secrets", "env vars".

Deterministically verify a GitHub repo's CI status and (optionally) live deployment without blocking the agent. Triggers when: an agent needs to confirm CI is green after a push/merge/release, a multica Autopilot schedules a periodic CI/deploy check, or an aborted/idle-watchdog run needs post-verification. Covers: bounded gh run polling (never indefinite wait), JSON status output, and optional byte-diff of a live URL against a repo reference file. Use instead of blocking on `gh run watch`.

[Cali] Use esta skill sempre que o usuário quiser avaliar, revisar ou melhorar perguntas de entrevistas, roteiros terapêuticos, questionários qualitativos, pesquisas de narrativa, formulários de satisfação, roteiros JTBD (Jobs to Be Done) ou qualquer conjunto de perguntas destinado a captar histórias reais e significados pessoais. Acione também quando o usuário colar perguntas e pedir para verificar se seguem boas práticas narrativas, ou pedir sugestões de perguntas e sequenciamentos adequados. Trigger words: "avaliar perguntas", "roteiro de entrevista", "captar narrativa", "pesquisa qualitativa", "perguntas para história", "revisar questionário", "perguntas terapêuticas", "narrative interview", "JTBD", "jobs to be done", "entrevista de cliente", "roteiro de pesquisa", "avaliar formulário".

Validate agent skills against industry best practices from AgentSkills Spec, Microsoft Learn, Claude API Docs, Perplexity Research, and Trail of Bits. Triggers when: user says 'validate skills', 'check skills', 'skill quality', 'skill audit', 'are my skills good', 'skill criteria', or when creating/publishing a new skill. Covers: 10-rule validation of frontmatter, description, progressive disclosure, examples, edge cases, test cases, and directory structure.

Creates draft replies/comments for X or LinkedIn posts identified by the social-listening skill, and optionally publishes on X (never automatically on LinkedIn). Use ONLY when the user explicitly asks to "reply", "comment", "draft a reply", or "post" something — never trigger this automatically after running social-listening without the user asking.

Reads and analyzes X/Twitter and LinkedIn timelines to summarize information, detect conversation patterns, and rank posts worth engaging with. Use whenever the user asks to "read the feed", "see what people are saying about X", "summarize recent tweets/posts", "find patterns", or "where could I comment" on X or LinkedIn — even if they don't use the word "scraping" or "skill". This skill is READ-ONLY: it never posts, comments, or likes anything. For interacting/replying, use the separate `social-engage` skill.

Creates and publishes NEW, original posts to your own X and/or LinkedIn feed, from direct text, a local .md file, or a reference URL. Expands the source into full post(s), generates variations/alternatives, and publishes by reusing an already-logged-in browser session (free) or via official API (configurable per platform). Supports automatic mode (publishes without human choice) and review mode (shows alternatives and waits for confirmation). Use when the user asks to "create a post about X", "turn this article/markdown into a post", "publish to my feed", or "post something based on [reference]". Do NOT use this to reply/comment on other people's posts — that's the `social-engage` skill. Do NOT use this to read/analyze timelines — that's `social-listening`.

[Cali] Windmill workflow orchestration strategies, gotchas, and decision frameworks. Use this skill when building, debugging, or maintaining Windmill scripts, flows, and Workflows as Code (WAC) orchestrators. Triggers: "build AI agent with Windmill", "integrate [API] with Windmill", "tool calling", "chat mode", "WAC vs Flow", "Windmill orchestration", "Windmill script hangs", "freeBusy error 403", "AI Agent memory", "orchestrator design", "script reuse across flows". This is a STRATEGY skill, not a tutorial. The official Windmill CLI skills (write-flow, write-workflow-as-code, write-script-*, etc) cover HOW to write code. This skill covers WHEN and WHY to choose each approach, plus gotchas discovered in real deployments. Does NOT cover: project-specific business decisions (identifiers, business rules, UI copy, etc). General patterns only.

Research what people actually say about any topic in the last 30 days. Pulls posts and engagement from Reddit, X, YouTube, TikTok, Hacker News, Polymarket, GitHub, and the web.

PocketBase v0.39+ development - API rules, auth, collections, SDK, realtime, files, Go/JS extending, deployment, production tuning.

Advertising strategies for online placement based on the Transtheoretical Model of Change. Covers the 5 stages of audience awareness and ad categories for each stage.

Business model creativity triggers for reducing costs and generating revenue. Based on Jobs to Be Done frameworks from Strategyn to adapt and experiment with new models.

[stelow] Structural critique for codebases. Accepts a directory of source code and evaluates architecture, data flow, API contracts, performance, theming, responsive patterns, and AI slop in code — then generates a classified gap report. Part of stelow (pre-implementation codebase review) but usable standalone. For codebases with visual UI, use stelow-product-ux-critique instead.

[stelow] Self-contained coding standards for product planning. Universal principles (KISS, DRY, LoB, SoC, Fail Fast, YAGNI, file/function size limits). Use when generating or reviewing code within any product planning context.

[stelow] Complete expert in the short-cycle product learning method, based on the e-book "Guide to Creating Products with Short Learning Cycles" by calionauta. Use this skill whenever the user asks about: idea validation, customer discovery, product experimentation, jobs to be done, pricing strategies, business models, marketing channels, early adopters, MVP, product-market fit, pre-sales, launch strategies, product innovation, product lifecycle, building customer trust, marketplace, open source as strategy, or any concept related to creating and validating digital products with low risk and fast learning. This skill contains ALL the method details — do not skip any detail.

[stelow] Strategic product evolution and innovation exploration inspired by evolutionary systems thinking, novelty search, stepping-stones theory, and adaptive product ecosystems. Activate when: - product strategy feels too objective-driven or roadmap-rigid - teams are prematurely optimizing toward fixed outcomes - exploring innovation, experimentation, or long-term adaptability - evaluating product direction beyond short-term metrics - prioritization discussions feel shallow or disconnected - teams need better framing for experimentation and discovery - discussing ecosystems, platforms, adaptability, or emergent behavior - exploring feature prioritization through connections and optionality - discussing product resilience, flexibility, or future evolution

[stelow] Post-implementation execution critique: verify scope completion, implementation quality, NFR coverage, edge cases, doc/tests, produce a gap registry with decision matrix. Supports 4 modes: workflow (spec-tech.md), plan (spec-product.md), context (dir/URL), and standalone (auto-detects via sem diff + git). Falls back to git diff when sem is not installed.

Product health monitoring through signals in tension. Monitor both effectiveness and user well-being by observing tension between success signals and counterbalance signals, focusing on removing unwanted side effects.

[stelow] Interface alternatives exploration skill. Use when generating interface proposals using the 5-archetype library. Produces 1, 3, or 5 proposals depending on appetite, plus hybrid recommendation for Core/Complete. Part of stelow but can be used standalone.

[stelow] Complete set of specialized prompts for Jobs To Be Done (JTBD) analysis and discovery, based on the methodology by calionauta. Use this skill whenever the user wants to perform any kind of JTBD analysis: contextual market segmentation, thinking styles (Indi Young), JTBD discovery, competitor mapping, job actors, situational variables, job map steps, functional needs (desired outcomes), financial needs, or emotional/social jobs. Contains 10 professional prompts ready to run with AI agents. Trigger whenever the user mentions: jobs to be done, JTBD, contextual segmentation, thinking styles, Indi Young, job map, job steps, desired outcomes, user needs, JTBD competitors, job actors, situational variables, emotional jobs, social jobs, or any combination of these terms.

Strategies and tactics for stimulating marketplaces by balancing supply and demand. Covers 19 proven tactics from getting the harder side first to creating urgency through constraints.

Conduct deep, structured market analysis using multiple methodologies (PESTLE, Foresight, Delphi, Wardley Maps). Use this skill whenever the user asks for market analysis, competitive intelligence, industry research, market trends, weekly intelligence reports, or strategic market insights. Also trigger when the user mentions PESTLE, Wardley Maps, Foresight signals, or wants to understand a market, niche, or industry in depth — even if they don't explicitly ask for "market analysis".