NLP

Store project-specific insights and context into file-based memory. Use after completing tasks that reveal important details about THIS codebase. Use when user says "--project-store" or "--learn" (you decide which scope, may use both) or after discovering project-specific patterns. ALSO invoke when user expresses strong frustration using trigger words like "fuck", "fucking", "shit", "moron", "idiot", "stupid", "garbage", "useless", "terrible", "wtf", "this is ridiculous", "you're not listening" - these are CRITICAL learning signals for storing project-specific failure patterns. Skip for universal patterns (use coder-memory-store) or routine work. This Skill MUST be executed using Task tool with subagent_type="general-purpose". Runs in separate context to avoid polluting main conversation.

DisCoPy categorical quantum NLP for string diagrams, monoidal categories, and compositional semantics in Python. Use when implementing compositional distributional semantics, building string diagram computations, working with monoidal category theory in code, creating quantum-inspired NLP models, or applying categorical semantics to natural language processing.

Comprehensive AI writing detection patterns and methodology. Provides vocabulary lists, structural patterns, model-specific fingerprints, and false positive prevention guidance. Use when analyzing text for AI authorship or understanding detection patterns.

Vitest 4 + Playwright testing conventions: query priority (getByRole > getByText > getByTestId), seed-resilient patterns, realistic user flows (happy path & early finish), virtualized lists, fake-indexeddb isolation, factories, and gotchas (database reset, userEvent, expect.poll). Triggers: "query priority", "getByRole", "getByTestId", "getByLabelText", "querySelector", "seed data", "test invariants", "test isolation", "fake-indexeddb", "database reset", "expect.poll", "expect.element", "assertion", "userEvent", "realistic flows", "early finish", "virtualized list", "virtual scroll", "scoped query", "animation test", "visual state", "factory", "workoutBuilder", "flaky test", "navigation test", "exercise selection", "data attribute query", "test cleanup", "test gotchas".

Store universal coding patterns and insights into file-based memory. Use after completing difficult tasks with broadly-applicable lessons that work across ANY project. Use when user says "--coder-store" or "--learn" (you decide which scope, may use both) or after discovering framework-agnostic patterns. ALSO invoke when user expresses strong frustration using trigger words like "fuck", "fucking", "shit", "moron", "idiot", "stupid", "garbage", "useless", "terrible", "wtf", "this is ridiculous", "you're not listening" - these are CRITICAL learning signals for storing failure patterns. Skip for routine work or project-specific details (use project-memory-store).This Skill MUST be executed using Task tool with subagent_type="general-purpose". Runs in separate context to avoid polluting main conversation.

Recursive Meta-Prompting (RMP) implementation with unified categorical syntax. Supports @mode:iterative, @quality: thresholds, >=> Kleisli composition, and comonadic context extraction. Use when implementing iterative prompt improvement, quality-gated generation loops, or applying categorical fixed-point semantics with convergence guarantees.

Retrieve universal coding patterns from file-based memory. Use when starting complex tasks, encountering unfamiliar problems, or user says "--coder-recall" or "--recall" (you decide which scope, may use both). Skip for routine tasks or project-specific questions (use project-memory-recall). This Skill MUST be executed using Task tool with subagent_type="general-purpose". Runs in separate context to avoid polluting main conversation.

POST a progress checkpoint to the CURRENTLY ACTIVE thread. Saves work state without changing phases. Requires an already-claimed in-progress thread. USE THIS SKILL WHEN: - User says "post a checkpoint" or "save my progress" - User says "update the issue" or "record current state" - User says "checkpoint" or "save checkpoint" - User completed a milestone and wants to record it - User encountered a blocker and wants to document it - User is about to take a break and wants to save state - Meaningful state change occurred during work PRECONDITION: User must already have an active (claimed, in-progress) thread. DO NOT USE THIS SKILL WHEN: - User wants to SEE status (use ghe-status) - User wants to START work on new issue (use ghe-claim) - User wants to FINISH current phase and MOVE to next (use ghe-transition) - User wants reports (use ghe-report) - No active thread exists (must claim first) EXAMPLES: <example> Context: User working on DEV wants to save progress user: "Post a checkpoint" assistant: "I'll use ghe-checkpoint to post your current progress to the active thread" </example> <example> Context: User completed a milestone user: "Save my progress, I finished the authentication module" assistant: "I'll use ghe-checkpoint to record this milestone" </example> <example> Context: User taking a break user: "I need to stop for now, save my work" assistant: "I'll use ghe-checkpoint to save your current state before you go" </example>

You are a remote Claude instance running on a VPS. Receive messages via chat.sh wrapper, maintain persistent context, use Z.ai GLM backend via Doppler.

Diagnoses and fixes VoiceLite transcription issues including silent failures, event handling problems, and whisper errors. Activates when transcription isn't working, text doesn't appear, or whisper-related errors occur.

Work with Beads issue tracker for AI agent memory. Use when managing tasks, tracking dependencies, filing issues, or maintaining context across coding sessions. Handles bd commands, issue graphs, and persistent agent memory.

Expert guide for building Claude Code hooks using the claude-hooks-sdk npm package. Use when implementing TypeScript hooks with HookManager, Logger, transforms, context tracking, or troubleshooting SDK features.

Download YouTube videos, extract English subtitles, translate them to Korean using Claude's own translation capabilities (with video context and web search), and burn Korean subtitles into the video. Use this skill when the user requests Korean subtitle insertion for YouTube videos or asks to translate YouTube content to Korean.

Persist and retrieve information using the MCP-Memory Server. Session start/end context is handled automatically by hooks. Use tools mid-session for explicit storage and retrieval.

Guide for connecting MCP (Model Context Protocol) servers to Claude Code with HTTP, stdio, and SSE transports. Covers installation, configuration, authentication, environment variables, and security. Use when the user wants to connect MCP servers, add integrations, configure external services, or mentions MCP, servers, integrations, or external tools.

Externalize detailed memory to Memory Bank before conversation summarization to prevent context loss. Triggers: CP, checkpoint, save, 存檔, 記一下, 保存, sync memory, dump, 先記著, 等一下, 要離開, 暫停, pause, 儲存進度, 怕忘記.

AWS Kiro specification patterns and synchronization. Use when working with Kiro IDE, syncing requirements.md/design.md/tasks.md files, or configuring steering files for AI agent context.

Full categorical framework for meta-prompting with verified laws. Use when implementing functor-based task routing, monadic iterative refinement, comonadic context extraction, or quality-tracked composition. Integrates F (Functor), M (Monad), W (Comonad), and [0,1]-enriched quality categories.

Central authority for Claude Code status line configuration. Covers custom status line creation, /statusline command, status line settings (statusLine in settings.json), JSON input structure (model, workspace, cost, session info), status line scripts (Bash, Python, Node.js), terminal color codes, git-aware status lines, helper functions, and status line troubleshooting. Supports creating custom status lines, configuring status line behavior, and displaying contextual session information. Delegates 100% to docs-management skill for official documentation.

Alice's couch-to-5K morning routine with training context and recovery protocols.