pragmatic

Plugin pragmatic from marketplace pragmatic-skills in sapoepsilon/pragmatic-skills. Commands read from the repo's manifest — 1 month ago.

Skills (14)

• backend-comprehensive-review

  Use when reviewing a backend/API PR end-to-end with a comprehensive multi-agent review: understand the service architecture, inspect the diff, review migrations and authorization, run available checks, locate the project-configured non-prod deployment, verify the changed endpoints with real requests

• loopcopy

  Turn what you are working on or brainstorming right now into one self-contained `/loop` command and put it on the clipboard, so it can be pasted into a fresh session on any machine (macOS, Windows, Linux, WSL, SSH) and run on a schedule. Use when the user says "loopcopy", "make me a loop command", "

• macos-screen-recorder

  Use when producing privacy-safe QA or product-demo evidence from a local or remote macOS machine. Stages a clean desktop, routes signing and GUI capture through the logged-in Aqua session, reviews a screenshot and sampled video frames before upload, publishes through user-local storage configuration

• mediator-loop

  Run a blind implement -> review -> fix loop between two agents that never learn of each other's existence, mediated until an independent reviewer returns APPROVED. Engine-agnostic — implementer and reviewer can be any two CLIs (or the same one twice). Use when the user says "mediator loop", "blind l

• mobile-analyze

  Stage 1 of the mobile auto-shipper — gather context from Linear, Fireflies, GitHub, Sentry, code, screenshots, backend/frontend repos, and connected devices to confirm exactly what the developer wants before implementation. Use when starting mobile/backend/frontend autoship work, or when a later sta

• mobile-comprehensive-review

  Use when reviewing a mobile app PR end-to-end with a comprehensive multi-agent review: understand architecture, inspect the diff, run available checks, build the app, verify behavior on the project-configured real device/emulator and non-prod backend, optionally record evidence, and produce a PR-rea

• mobile-implement

  Stage 2 of the mobile auto-shipper — implement a confirmed mobile/backend/frontend change yourself on the exact branch, then build/test until green before QA. You are the coder. Use after mobile-analyze approval or when mobile-qa bounces a precise failure back.

• mobile-qa

  Stage 3 of the mobile auto-shipper — drive the real app/API on a real non-prod target (iOS/Android simulator, physical device, browser, CLI/API), verify backend/Supabase integration, record evidence when possible, and either pass to ship or bounce a precise failure back to implementation. Use after

• mobile-setup

  Set up the mobile auto-shipper on this machine and in this project, then run a preflight that proves the pipeline works against the real repo. Use when the user says "set up mobile autoship", "mobile-setup", "configure autoship here", or wants mobile, backend, Supabase, and Flutter QA wiring. Probes

• muchotexto

  Answer in the minimum number of sentences that still carries the meaning. Start at exactly one and proactively grow by +1 whenever a fact, caveat, or required step would otherwise be lost — never pad, never stop short. Use when the user says "muchotexto", "mucho texto", "tldr", "too long", "shorter"

• resume

  Resume a Claude Code session by pulling recent session logs and project decisions from an Obsidian vault exposed via an MCP filesystem server. Use when the user types /resume, says "resume", "where did I leave off", "what was I doing", "continue from last session", or otherwise asks you to reload co

• review-fix-loop

  Address external review findings (PR review comments, audit reports, QA sweeps) through an orchestrated watcher → implementer → verifier loop with model-pinned subagents. Use when the user says "address these review comments", "fix what the reviewer found", pastes a review URL, or wants findings fix

• save

  Save the current Claude Code session as a dated log in an Obsidian vault exposed via an MCP filesystem server. Use when the user types /save, says "save this session", "save my progress", "log this", "end of session", or otherwise asks you to persist what was done so a future session can resume it.

• verify-to-e2e

  Verify a new or changed feature by driving the real running app (iOS/Android simulator, browser, CLI, or API) via MCP or the project's own automation, then codify that exact check into a reusable end-to-end test wired into the project's existing test harness. Use after building something you were as

Similar plugins