lakr233
- 7 skills
- 0 followers
- 18 hours ago last updated
- ▌ Code Clarity · lakr233 bundleWrite, review, and simplify Go, Swift, Objective-C, and TypeScript code so behavior has one owner, state is minimal and derivable, failures are grouped by real recovery behavior, APIs expose only necessary concepts, control flow stays flat, and tests protect observable outcomes without forcing production abstractions. Use when code grows through incremental feedback, a diff adds many states/errors/branches/types, naming or flow is hard to follow, calls, conditions, or declarations are mechanically wrapped to a column count, seams and mocks multiply, or a refactor should reduce concepts while preserving behavior and repository conventions.
- ▌ Website Login · lakr233Use whenever the user needs to log in to a website on their phone and reuse that authenticated session locally, especially for Playwright storageState JSON, cookies/localStorage capture, MFA, SMS codes, passkeys, or mobile-first login flows that are awkward in headless automation. You run Cookey CLI in your current environment; the user signs in on their iPhone in the Cookey app; the relay only transports encrypted blobs. Trigger even if the user asks more loosely for help logging in, reusing an authenticated browser state, exporting cookies, or getting a local automation session from a phone login.
- ▌ Dribbble Reference Harvester · lakr233 bundleUse when the user wants to collect design references from Dribbble into a local folder, build a reference set for AI design prompting, or run the Dribbble CLI in this repo with a search query, limit, output directory, or headful debugging.
- ▌ Flowdown Agent · lakr233Use when helping users evaluate, install, configure, automate, or troubleshoot FlowDown AI, a privacy-first native AI workspace for iOS, iPadOS, and macOS with local models, OpenAI-compatible providers, MCP client support, web search, attachments, memory, and Apple Shortcuts automation.
- ▌ New Model Hunt · lakr233 bundleWhole-repository adversarial review and patch run for trying out a new model: parallel correctness hunters and code-clarity reviewers per subsystem plus cross-cutting lenses, one skeptic per finding prompted to refute it, a consolidated report saved to the Desktop, then per-subsystem patch agents, build/test, one commit per subsystem, a UI-copy pass, and (on request) the release. Use when the user says /new-model-hunt, "run the hunt with the new model", "hunt bugs in the whole repo", "adversarial review", or wants to benchmark a new model on this codebase. Every judgement is anchored, refuted-by-default, and forbids over-guarding and over-design.
- ▌ UI Copy Polish · lakr233 bundlePolish user-facing UI copy in any codebase (Xcode, npm/Node, Rust, Go, Python, Android, Flutter, .NET, docs) with a workflow of subagents — find → verify → fix → review → fix (loop) → localize → commit → push. Use when the user runs /ui-copy-polish, or asks to 过一遍文案, polish UI copy, rewrite user-facing strings, 书面化, make copy UI-facing, or fix informal / developer-facing labels, errors, empty states, and alerts.
- ▌ Kernel Analysis Vphone600 · lakr233 bundleAnalyze vphone600 kernel artifacts using the local symbol database and XNU source tree. Use when working on kernel reverse engineering, address-to-symbol lookup, release-vs-research kernel comparison, or patch analysis for vphone600 variants in this repository.