Afegits recentment
Els últims skills afegits al catàleg. Descobreix les novetats!
dotfiles-docs-and-writing
taloncjones
Use when writing or updating any doc of record in this dotfiles repo -- README.md, repo CLAUDE.md, claude/global-instructions.md, claude/operating-principles.md, codex/AGENTS.md, NOTICE -- or when authoring a new skill or slash command file. Trigger on "update the README", "document this", "where should this fact live", "add a skill", "add a command", "the docs are stale", "docs drift", or any house-style question (emoji tags, dash style, commit-message format). Not for platform mechanics content itself (use claude-code-platform-reference), change gating and review process (use dotfiles-change-control), or public-repo/licensing posture decisions (use dotfiles-external-positioning).
dotfiles-diagnostics-and-tooling
taloncjones
Use when you need to MEASURE the state of a dotfiles machine or cloud container instead of eyeballing it -- interpreting identity-doctor output, running/reading bin/dotfiles-tests, inspecting plugin state (installed_plugins.json, known_marketplaces.json), diffing live settings.json against the template, or running this skill's cloud-doctor.sh / symlink-audit.sh scripts. Trigger phrases -- "is this environment healthy", "why did identity-doctor fail", "which test suite broke", "are the plugins actually installed", "did settings drift", "audit the symlinks". NOT for fixing what you find (dotfiles-run-and-operate), rebuilding from scratch (dotfiles-build-and-env), debugging a specific symptom (dotfiles-debugging-playbook), or single-fact verification recipes (dotfiles-verification-toolkit).
dotfiles-debugging-playbook
taloncjones
Symptom-to-fix triage for this dotfiles repo. Load when something is BROKEN and you need to diagnose it -- plugin skills/commands missing from a session, settings.json truncated, dangling or write-through symlinks, "git refuses to commit" or wrong commit identity, SSH offering the wrong key, a guard hook blocking a commit, a cloud session missing its config, blank statusline, dead hooks, zsh nags, prettier rewriting files unexpectedly, or a worktree missing .todos/.planning. Trigger phrases: "why is X missing", "commit blocked", "hook didn't run", "wrong identity", "symlink broken", "plugin not installed". NOT for routine operation (dotfiles-run-and-operate), proving healthy state from first principles (dotfiles-verification-toolkit), or the history of past investigations (dotfiles-failure-archaeology).
dotfiles-config-and-flags
taloncjones
Catalog of every environment variable, script flag, and config knob in this dotfiles repo -- consumer, default, effect, and whether it is production, test-only, or emergency-bypass. Load when asking "what does VAR do", "is there a flag for X", "how do I skip a hook", "what reads CLAUDE_CONFIG_DIR / CLAUDE_CODE_REMOTE / GSD_HOOK_REPAIR" (the retired ECC's ECC_VENDOR_LANGS no longer exists), "why is --local ignored", "what keys does settings.json.tmpl own", or before adding a new flag. NOT for diagnosing breakage (dotfiles-debugging-playbook), running the update/plugin lifecycle (dotfiles-run-and-operate), platform semantics of the vars Claude Code itself defines (claude-code-platform-reference), or recreating a machine (dotfiles-build-and-env).
dotfiles-cloud-first-campaign
taloncjones
The executable campaign runbook for making cloud/ephemeral Claude Code sessions (claude.ai/code containers) a first-class primary environment for this dotfiles setup. Load when planning or executing cloud-parity work -- "make cloud sessions first-class", "roll out plugins to another repo's cloud environment", "why aren't plugins loaded on session 1", "should the Setup script or the settings declaration carry the install", "close the machine/cloud parity gap", "was ECC rules vendoring still needed before ECC's own retirement". Phased with decision gates, exact commands, expected outputs, and fenced wrong paths with incident hashes. NOT for diagnosing a single sick container (dotfiles-diagnostics-and-tooling cloud-doctor), one-time environment rebuild (dotfiles-build-and-env), the settled incident history itself (dotfiles-failure-archaeology), or Claude Code platform mechanics in general (claude-code-platform-reference).
dotfiles-change-control
taloncjones
Load BEFORE making, committing, or reviewing ANY change in the dotfiles repo. Covers change classification (tracked asset vs installer logic vs machine-local template vs vendored-untracked vs project .claude config), the full gate chain (guard hooks, bin/dotfiles-tests, public-safety scan, CI), commit/branch/PR rules, and the non-negotiables with their incident history. Trigger phrases -- "commit this", "open a PR", "why did my commit get blocked", "can I edit settings.json.tmpl", "can I edit these rules files", "hook blocked me", "bypass the hook". NOT for diagnosing why something is broken (dotfiles-debugging-playbook), running the tests as evidence (dotfiles-validation-and-qa), or historical why-decisions (dotfiles-failure-archaeology).
dotfiles-build-and-env
taloncjones
Recreate any dotfiles environment from nothing -- fresh macOS, fresh Linux, claude.ai/code cloud container, or the work-identity add-on -- as numbered copy-paste runbooks with expected output at each checkpoint, plus the twelve known install traps. Load when the user says "set up a new machine", "fresh install", "bootstrap", "new laptop", "cloud container setup", "install from scratch", "add work identity", or when bootstrap.sh / bootstrap-cloud.sh / identity-setup fails mid-run. NOT for day-to-day update/plugin operations on an already-installed machine (dotfiles-run-and-operate), diagnosing a broken existing install (dotfiles-debugging-playbook), or proving a link/plugin/identity claim (dotfiles-verification-toolkit).
dotfiles-architecture-contract
taloncjones
Load BEFORE proposing any structural or design change to the dotfiles repo, and whenever you catch yourself asking "why is it built this way", "can I just track this file", "why is settings.json not symlinked", "why two config dirs", "why is this untracked", or "can the SessionStart hook install the plugin". Documents the eight load-bearing design decisions with rationale, the invariants that must hold, and the known weak points, plus the install-layout reference (install flow, symlink and seed targets, hook catalogue) and the Codex plugin integration reference (codex-surfaces.py / codex-roles.py reconciliation, staging provenance). NOT for making/committing a change (dotfiles-change-control), diagnosing a breakage (dotfiles-debugging-playbook), or the blow-by-blow incident history (dotfiles-failure-archaeology).
claude-code-platform-reference
taloncjones
Load when you need Claude Code PLATFORM mechanics as used by this repo -- config-dir resolution and CLAUDE_CONFIG_DIR, what lives in ~/.claude, settings.json precedence and key ownership, plugin/marketplace internals (installed_plugins.json, known_marketplaces.json, marketplace manifests, CLI verbs), hook events/matchers/exit codes, skills vs commands vs agents, CLAUDE.md @import, cloud-container env vars and launch ordering, statusline wiring. Trigger phrases - "how do plugins actually install", "where does settings.json come from", "what is CLAUDE_CONFIG_DIR", "why did my hook not fire", "what runs at session start", "what is installed_plugins.json". This is the factual reference; for step-by-step operations use dotfiles-run-and-operate, for env recreation use dotfiles-build-and-env, for proving state use dotfiles-verification-toolkit.
triagem
dspinillo
Julga vagas pendentes do Radar de Vagas e aprende pontos de atenção com descartes, sem esconder ou descartar vagas.
preencher-formulario
dspinillo
EXPERIMENTAL. Preenche no navegador o formulário de candidatura de uma vaga do Radar a partir da folha de respostas, pelo Claude in Chrome, e para antes do envio.
onboarding
dspinillo
Entrevista em português para iniciar ou atualizar o Radar de Vagas, salvar perfil e busca e revisar currículo e LinkedIn.
entrevista
dspinillo
Registra uma entrevista informada pela pessoa e monta uma página de preparo por vaga com casos reais, lacunas e perguntas.
curriculo-por-vaga
dspinillo
Monta um currículo HTML imprimível e legível por ATS para uma vaga do Radar, usando somente fatos do perfil.
candidatura
dspinillo
Monta a folha de respostas de uma vaga do Radar, pronta para revisar e colar no formulário, com alertas, personalização da Gupy e respostas comuns, usando só fatos do perfil e dos dados de candidatura.
acompanhamento
dspinillo
Atualiza etapa, próximo passo e observações de um processo no Radar quando a pessoa conta uma novidade.
social-growth
8-Sync-Dev
OPT-IN (không auto-bật). Use ONLY when the user explicitly enables this skill or asks to build social-media marketing / branding / lead-gen for 8 Sync Dev — campaigns on Facebook, YouTube, TikTok (and Fanpage / community groups), setting up a Page properly, audience insight, monthly content plan + targets/KPIs, or finding leads & customers. Composes with assp-skill (brand voice + avatar + offer) and last30days (recency research); never replaces them. Produces a concrete monthly plan with per-channel targets, content calendar, funnel, and lead-capture mechanics — never generic "post consistently" advice.
senior-security
8-Sync-Dev
Security engineering toolkit for threat modeling, vulnerability analysis, secure architecture, and penetration testing. Includes STRIDE analysis, OWASP guidance, cryptography selection tables, and a map of the security scanners to reach for.
senior-frontend
8-Sync-Dev
Frontend development skill for React, Next.js, TypeScript, and Tailwind CSS applications. Use when building React components, optimizing Next.js performance, analyzing bundle sizes, scaffolding frontend projects, implementing accessibility, or reviewing frontend code quality.
research-paper
8-Sync-Dev
Work a paper, claim, or benchmark end to end — replicate a result, extract an implementable ML training recipe, audit a paper against its released code, draft a paper-style write-up, or run a bounded try/measure/keep-or-revert experiment loop. Use when the user wants to reproduce, verify, optimize against, or formally write up research rather than get a casual summary.
remote-compute
8-Sync-Dev
Run research, replication, or benchmark code somewhere other than the bare host — a Docker sandbox, Modal serverless GPUs, or a persistent RunPod GPU pod. Use when the user picks an isolated/remote execution environment, or asks to run untrusted or GPU-heavy code safely.
ponytail
8-Sync-Dev
Forces the laziest solution that actually works, simplest, shortest, most minimal. Channels a senior dev who has seen everything: question whether the task needs to exist at all (YAGNI), reach for the standard library before custom code, native platform features before dependencies, one line before fifty. Supports intensity levels: lite, full (default), ultra. Use whenever the user says "ponytail", "be lazy", "lazy mode", "simplest solution", "minimal solution", "yagni", "do less", or "shortest path", and whenever they complain about over-engineering, bloat, boilerplate, or unnecessary dependencies.
nextjs-app
8-Sync-Dev
Use when building, editing, or debugging a Next.js App Router app (Next 15/16, React 19) — anything involving the app/ route tree, Server Components vs 'use client', Server Actions, Route Handlers, the fetch cache, middleware.ts, SEO/metadata, an Encore.ts (or other) backend client, a Turborepo monorepo, or Vercel/Docker deployment. Produces correct RSC boundaries, non-stale caching, verified UI. Grounded in the Next.js 16 App Router canon; extends with the 8syncdev repo patterns + omp-native tooling (codegraph · codebase-memory · serena · engine_* · browser).
locate-anything
8-Sync-Dev
Fast open-vocabulary VISUAL GROUNDING via NVIDIA LocateAnything-3B (run through the mudler/locate-anything.cpp ggml port, no Python). Turns an image + a text description into labeled bounding boxes + click-center coordinates. Use when you need EXACT PIXEL COORDINATES from an image — GUI element grounding (to click a button/field precisely), OCR/text localization, open-set object detection — rather than a prose description. Invoked as 8sync locate <image> "<target>". Complements a native VLM / zai-vision (which describe) and the browser tool (which clicks).