Skills
Persistent memory for Claude Code
Provides Claude Code with persistent memory across sessions by extracting durable lessons, decisions, and bug fixes into markdown files. It enables the retrieval of relevant context at the start of each new session.
Persistent memory and context management for LLMs
This skill provides persistent memory across sessions by mining completed interactions to extract durable lessons. It pre-loads relevant context at the start of new sessions, eliminating the need for repeated project explanations.
Automated Claude Code Session Knowledge Extraction
A background daemon that automatically extracts knowledge from completed Claude Code sessions. It polls for new session transcripts and processes them after they have concluded.
Check detailed memem memory system status
Executes a dedicated command to retrieve comprehensive operational metrics for the memem memory system. This includes reporting memory capacity, active projects, search database status, and the current health of the associated miner.
Continuous regression monitoring during development
This skill provides continuous regression monitoring by observing specified directories for file changes and automatically re-running evaluation checks. It displays a live scorecard, allowing developers to track pass/fail status and score d…
Mining Claude Code Session History
Extracts knowledge from all historical Claude Code sessions, including those predating installation. It also initiates a background daemon to automatically mine new sessions as they occur.
Memem System Preflight Health Check
Performs a preflight diagnostic to verify the integrity of the memem installation, checking for missing dependencies, environment variable configurations, and directory writability. It identifies critical failures or degraded states caused …
AI agent regression testing with EvalView
Detect regressions in AI agent behaviour by comparing current outputs and tool calls against golden baselines. It identifies changes in outputs, tool usage, and significant score drops.
Automated AI Agent Test Case Generation
Automate the creation of test suites for AI agents by generating cases from SKILL.md files, manual descriptions, or by capturing live agent interactions via a proxy.
Automated Code Review and Security Analysis
This skill automates the identification of security vulnerabilities, performance bottlenecks, and code quality issues, prioritising findings by severity.
Simple Greeting File Generator
This skill generates a text file named hello.txt containing a personalised greeting and the current date.
Comprehensive Code Review Assistant
An automated assistant designed to perform multi-level code analysis, focusing on security vulnerabilities, performance bottlenecks, and adherence to best practices. It generates structured review reports and suggests actionable improvement…
Firefox Browser Automation and Debugging
Automate Firefox browser interactions, perform E2E testing, and scrape web content using DOM snapshots and element UIDs. It also provides capabilities for monitoring console messages and network requests for debugging.
Specialized Korean web search via Naver MCP
This skill provides comprehensive search capabilities across the Korean web, accessing specialized Naver services for news, local listings, e-commerce, and academic content. It is ideal for tasks requiring deep integration with Naver's spec…
Semantic Code Analysis and Refactoring Toolkit
This skill provides access to JetBrains IDE indexing capabilities, enabling deep semantic code operations such as finding references, resolving definitions, and performing safe refactoring. It is designed to supersede basic text search for …
7-Layer Recursive Agent Memory Service
This service provides a robust, 7-layer recursive memory architecture for agents, supporting context branching and structured knowledge storage. It features a 5-pass compiler and ensures permanent, encrypted persistence via on-chain storage…
YouTube Resource Management via yutu CLI
Manage YouTube resources including videos, playlists, and comments using the yutu CLI. It interfaces with the YouTube Data API to perform operations such as uploading content, managing channel metadata, and moderating community interactions…
Evaluates and improves other agent skills
This skill assesses the quality of another agent's output against task requirements, providing structured JSON feedback. It identifies deficiencies, assigns a score (0.0 to 1.0), and specifies any missing instructions for improvement.
Evaluates politeness of pull request comments
This skill assesses the tone and politeness of a pull request review comment. It is designed to judge the communication quality, deliberately excluding technical correctness or code risk assessment.
Concise explanation of code differences
This skill provides a brief, high-level explanation of changes introduced within a given code diff. It is useful for quickly assessing the scope and potential impact of small code modifications.
Himalayas AI Hiring Assistant
An AI assistant for managing remote recruitment on the Himalayas marketplace. It enables users to post jobs, source talent, manage company profiles, and access market salary intelligence.
Batch analysis of architectural blueprints
This skill facilitates the batch analysis of up to 500 architectural blueprints via a dedicated API endpoint. Users can specify the analysis type, selecting from options such as 'full', 'dimensions', 'rooms', or 'materials' for comprehensiv…
Automated monorepo package publishing and release
This skill automates the entire monorepo publishing lifecycle, handling version bumping, git tagging, and building. It ensures deployment across all supported registries (e.g., npm, jsr) by managing state detection and triggering GitHub rel…
CCG Latent Chain Mode
A structured workflow for executing complex, multi-step development tasks through a four-phase reasoning process. It optimises token usage by employing hidden-state reasoning and transmitting only context deltas.