agentic-kit

Plugin agentic-kit from marketplace agentic-kit in timothyrusso/agentic-kit. Commands read from the repo's manifest — 1 month ago.

Skills (14)

• agent-device-configuration

  Install and configure agent-device, the mobile app verification toolkit the QA agents drive iOS and Android simulators, emulators and physical devices with. Registers the MCP server and sets the iOS runner environment variables. Run once per developer machine.

• agent-device

  Automates Apple-platform apps (iOS, tvOS, macOS), Android devices, and Amazon Vega OS TV apps in Vega Virtual Devices. Use when navigating apps, taking snapshots/screenshots where supported, driving TV remotes, tapping, typing, scrolling, extracting UI info, collecting evidence, or planning agent-de

• docs-audit

  Read-only audit of an app's documentation and Claude Code configuration for staleness against the current codebase and the agentic-kit plugin. Produces a report of what is outdated (stale or suspect) with the file and the fix. Use to check whether the app's docs, CLAUDE.md, kit.config.json and .clau

• dogfood

  Systematically explore and test a mobile app on iOS/Android with agent-device to find bugs, UX issues, and other problems. Use when asked to dogfood, QA, exploratory test, find issues, bug hunt, or test this app on mobile.

• grill-deck

  Harvest transferable engineering concepts and design rationale from the CURRENT session's transcript and write them as self-contained study flashcards to flashcards.md in the project root. Use only when the user explicitly runs /grill-deck (typically right before closing a session) to turn what was

• grill-me

  A relentless interview to sharpen a plan or design.

• grilling

  Grill the user relentlessly about a plan, decision, or idea. Use when the user wants to stress-test their thinking, or uses any 'grill' trigger phrases.

• handoff

  Compact the current conversation into a handoff document for another agent to pick up.

• implement-issue

  Implement a GitHub feature issue end to end. Judges the issue (grills the user only if something genuinely needs clarifying, folding answers back into the issue body), announces its reading, launches the agentic-kit:implement-issue-pipeline workflow (explore, build, wire PR, review, runtime QA, boun

• qa-baseline

  The standing baseline QA checks to run for EVERY feature, whatever it does, as a regression safety net, with the app-specific steps (setup, launch, first screens, navigation) read from kit.config.json. Run them on the device with agent-device before any feature-specific acceptance-criteria tests. Us

• teach

  Teach the user a new skill or concept, within this workspace.

• triage-pr

  Iteratively triage AI review-bot comments on an open pull request until the bots go quiet or a tripwire hands the loop back, vet each finding with finding-vetter, auto-fix confirmed ones, resolve noise with short replies, and consult the user in chat for judgment calls. Use when the user asks to tri

• write-issue

  Author a complete, well-specified GitHub feature issue by interviewing the user until the Description and Acceptance criteria are unambiguous, then create it with the Feature template structure and put it on the board named in kit.config.json. Use when starting a new feature from a rough idea; it fr

• writing-great-skills

  Reference for writing and editing skills well - the vocabulary and principles that make a skill predictable.

Subagents (6)

• code-reviewer

  Static code reviewer for a feature branch of an app built on the agentic-kit architecture. Reviews the diff against the kit's architecture and Effect rules and general correctness BEFORE runtime QA, focusing on what Biome, ESLint and dependency-cruiser do NOT catch. Read-only; returns blocking and n

• explorer

  Read-only pre-implementation explorer for an app built on the agentic-kit architecture. Given a feature issue, maps it onto the architecture (target feature and tier, the exact files and layers to touch, the closest existing pattern to mirror, integration points, risks) and returns a structured repo

• feature-builder

  Implements a feature from a GitHub issue in an app built on the agentic-kit architecture. Reads the issue, explores the relevant code, writes the implementation following the kit's architecture and Effect rules, opens a PR with small layer-aligned commits, and reports what it changed. Use to impleme

• finding-vetter

  Adversarial verifier for a single blocking finding from the implement-issue pipeline (code review or runtime QA) or from bot triage. Tries to REFUTE the finding against the actual diff, code and captured QA evidence before it may trigger an auto-fix round. Read-only; never edits code, never posts co

• qa-engineer

  Runtime QA specialist for the MOBILE targets (iOS or Android device, simulator, emulator) of an app built on the agentic-kit architecture. Drives the app through the agent-device CLI to verify a feature branch actually works. Runs the baseline checks plus the issue's acceptance criteria, captures sc

• qa-web-engineer

  Runtime QA specialist for the WEB targets (web Storybook, expo start --web, any browser-served surface) of an app built on the agentic-kit architecture. Drives a real Chromium through the agent-browser CLI to verify a feature branch works in the browser. Runs the web baseline checks plus the issue's

Similar plugins