jorgemenadev
- 32 skills
- 0 followers
- 7 hours ago last updated
- ▌ Visual Plan · jorgemenadev bundleTurn ordinary text plans into rich interactive visual plans with diagrams, file maps, annotated code, open questions, and UI/prototype review when useful.
- ▌
- ▌ Shiploop · jorgemenadev bundleTurn a plan, fix, or feature into a fully autonomous shipping run - GitHub issue ledger, dependency-gated worker tasks, gated phase PRs, and a reviewed final PR waiting for the human.
- ▌ Andy Post · jorgemenadev bundleDraft and publish social posts through Andy MCP. Use when the user wants to post, tweet, schedule, or create an X, LinkedIn, or Instagram post; when they give a raw idea for a social post; or when they run /andy-post.
- ▌ Counsel · jorgemenadev bundleConvene a two-vendor counsel — flagship reviewers from different vendors at high effort — to adversarially review an important architecture/design decision or proposal before committing. Use when the user explicitly asks for counsel or a two-vendor adversarial review.
- ▌ Research · jorgemenadev bundleInvestigate a question against high-trust primary sources and capture the findings as a Markdown file in the repo. Use when the user wants a topic researched, docs or API facts gathered, or reading legwork delegated to a background agent.
- ▌ Agentmail · jorgemenadev bundleOperate the agent's own AgentMail email inboxes via the `agentmail` CLI — send, receive, and poll for verification links/codes and invitations. Use when a task needs an agent-owned email address (service signup, auth-email verification, invite-flow testing, agent↔human mail), or when another skill's flow must receive or assert on an email.
- ▌ Grill With Docs · jorgemenadev bundleGrilling session that challenges your plan against the existing domain model, sharpens terminology, and updates documentation (CONTEXT.md, ADRs) in the target repo inline as decisions crystallise. The target repo defaults to the current one but may be a different repo the plan is meant for. Use when user wants to stress-test a plan against their project's language and documented decisions.
- ▌ SEO Growth Workspace · jorgemenadev bundleUse when starting, auditing, or operating SEO or organic growth for a product or local-business website — technical SEO, Search Console, keyword/content ops, schema, pSEO, local SEO/GBP, backlinks, AI-search visibility, conversion, and monthly reporting. Triggers: "set up SEO", "audit my site", "my traffic dropped", "why am I not ranking", "Search Console opportunities", "monthly SEO report", "how do we show up in ChatGPT/AI search". Creates a durable .seo workspace, captures business context, audits live/code/admin evidence, prioritizes a backlog, implements one high-leverage action, verifies reality, and logs handoff notes. Installs in a single site repo (standalone) or in an orchestrator/agent-profile repo managing many sites (hub). For standalone copywriting, paid channels, or email, use a dedicated skill.
- ▌ Orchestrate · jorgemenadev bundleOrchestrate multi-slice work as the conductor — plan slices with blocking edges, dispatch isolated executors from the frontier, review hand-backs, integrate, and verify to completion. Use when a task spans parallel agents, worktrees, AFK runs, shared resources, cross-runtime delegation, or an issue chain that must be driven to verified completion.
- ▌ Crew Dispatch · jorgemenadev bundleCrew-first dispatch for Matias — run a deliverable task through a crewmate with a durable crew record (intake → brief → lane → record → spawn → supervise → verify → integrate → close). Use when starting any multi-file change, investigation report, or product edit, or when reconciling crew/ records at session start, or when a dispatched crew must report completion after its launching turn ends. Inline-lane work (true one-liner, same-pass vault/tracker/home write, read-and-think answer) skips this skill.
- ▌ Posthog Growth Workspace · jorgemenadev bundleUse when operating data-driven growth for a product instrumented with PostHog — funnels, activation, retention, experiments, session-replay mining, campaign plays, and periodic growth reviews. Triggers: "growth review", "why aren't users activating", "run an experiment", "mine session replays", "what does the product data say", "set up the growth workspace", "monthly growth report". Creates a durable .growth workspace per product, reads live PostHog data CLI-first, does one verified high-leverage action per pass, and logs handoffs. Installs in a single product repo (standalone) or in an orchestrator/agent-profile repo managing many products (hub). Install/instrumentation doctrine and SEO/search-channel work belong to other skills.
- ▌ Writing For Agents · jorgemenadev bundleWriting documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.
- ▌ Session Wrap Up · jorgemenadevWrap up the session and update all documentation. Use when the user says 'wrap up', 'update the documentation', or is ending a work session.
- ▌ Operator Handoff · jorgemenadev bundleJob-file handoff between a requesting agent and the Operator — a human-supervised computer-use agent driving the machine's real desktop and browser. Use when the user explicitly requests the Operator, resumes an Operator run ("check the operator report", "run job NNN"), or a native browser/computer-use run reaches an actual human-only gate.
- ▌ Source To System · jorgemenadev bundleTurn external material into system improvements. Use when a user shares a source and asks whether to adopt it or what it should change in an existing workflow, repository, knowledge base, product, or operating system.
- ▌ Afk Pipeline · jorgemenadev bundleOperate the AFK pipeline for a registry-confirmed repository. Use when the user explicitly asks for AFK work, Pipeline Flags, an Agent Brief, pipeline dispatch, or pipeline installation, or when the active workspace contract routes substantial eligible development through AFK.
- ▌ Andes Release · jorgemenadev bundleTag and publish GitHub Releases for Andes products so the repo's release history matches the stores and production. Separate iOS and Android version lines plus web. Use when a production build finishes, Apple or Play confirms a version live, a batch of web PRs is worth naming, a repo has no release workflow yet, or someone asks which version is live.
- ▌ Storage Audit · jorgemenadevReclaim disk on Jorge's Mac mini with scripts/storage-hygiene.sh — worktrees, build and dependency caches, T3 Code thread history, Xcode data, tool caches. Use when free space is low, Jorge asks to clean up space, or the storage-hygiene cron needs diagnosis.
- ▌ Work Tracking · jorgemenadevUse when a multi-step plan needs durable tracking across a repo: markdown plans, GitHub Issues, Linear, memory, or a mix. Helps decide where work should live, when to promote tasks to issues, and how future agents should find the current state.
- ▌ Grok Deep Research · jorgemenadev bundleUse when the user wants iterative web research, asks Grok to investigate a topic deeply, or requests a bounded research loop. Runs Grok Build CLI in an isolated runtime, defaults to native web search, optionally enables only Firecrawl MCP, and produces a stable cited report plus an iteration ledger.
- ▌ Product Feedback Report · jorgemenadev bundleCreate and resume founder-led product feedback reports for web or mobile software. Use when someone wants to record product problems or ideas over time, continue an unfinished testing report, attach screenshots or short videos, or export a polished PDF for a product team.
- ▌ Cursor Subagent · jorgemenadev bundleCursor sidecar delegation. Use when explicitly asked to spawn Cursor or a Cursor-hosted Grok model for exploration, review, or isolated implementation. Native xAI Grok Build CLI requests belong to grok-cli-runtime.
- ▌ Domain Modeling · jorgemenadev bundleBuild and sharpen a project's domain model. Use when the user wants to pin down domain terminology or a ubiquitous language, record an architectural decision, or when another skill needs to maintain the domain model.
- ▌ Grok CLI Runtime · jorgemenadevxAI Grok auth and quota doctrine for any harness that runs the `grok` binary — T3 Code's `grok` provider instance, a headless `grok -p` sidecar, or a cloud runner seeded from GROK_AUTH_B64. Use when a Grok run fails with 401, 402, or 403, when Grok output is empty or DOA, or before any `grok login`.
- ▌ Sync Github Fork · jorgemenadev bundleKeep a formal GitHub fork current with its parent without rewriting published history. Use when adding or checking an upstream remote, measuring fork divergence, syncing upstream changes into a fork's default branch, or refreshing a local checkout after the fork updates.
- ▌ Codex CLI Runtime · jorgemenadevClaude-side contract for calling an installed Codex companion runtime. Use when Claude Code delegates a task, review, question, or implementation to Codex/OpenAI.
- ▌ Claude CLI Runtime · jorgemenadevClaude Code headless runtime contract. Use when a non-Claude runtime delegates one task, review, question, or one-vendor second opinion to Claude/Fable. Two-vendor adversarial review belongs to counsel.
- ▌ Mobile Monorepo IOS · jorgemenadev bundleBuild, debug, and release mobile apps across Expo/React Native, native iOS, native Android, or bounded hybrid architectures. Use for development builds, device proof, native-runtime changes, release workflows, EAS, TestFlight, App Store Connect, Google Play, or cross-platform mobile delivery.
- ▌ Using Git Worktrees · jorgemenadev bundleUse when creating an isolated Git workspace, running the default branch locally, or closing development work that used a worktree.
- ▌ Design System Keeper · jorgemenadev bundleExtract, encode, and enforce a repo's real design system as an agent-readable canon (v0 Design Systems 2.0 shape). Use when the user wants agents to learn or import their design system, asks why pages look inconsistent or like different design systems mixed together, wants a design consolidation or cleanup, or when building UI in a repo that has a canon (`design/` + DESIGN.md).
- ▌ Opencode CLI Runtime · jorgemenadevRuntime contract for delegating work to OpenCode CLI as a headless sidecar. Use when a non-OpenCode runtime delegates one task, review, question, or one-vendor second opinion to OpenCode. OpenCode's own subagents are native delegation; this is the cross-runtime contract.