Claude Skills

The internal-comms skill provides Claude with company-specific templates and guidelines for writing various internal communications. It should be used when creating devlog reports, leadership updates, project status reports, incident reports, and formatted updates like 22A/22B. The skill ensures all internal communications follow standardized formats by loading appropriate templates from its examples directory.

The Goose Recipes skill enables developers to create and validate reusable AI agent configurations for Claude. It helps with building recipe files that package agent setups including extensions, parameters, retry logic, and structured outputs. Use this skill when configuring Goose workflows or debugging recipe validation errors.

The goose-recipe-analysis skill enables developers to create Goose recipes for document analysis and transformation tasks using existing data files. It's designed for analyzing saved reports, transforming markdown files, and extracting insights without requiring MCP server access or live data queries. Use this skill when working with previously generated content rather than querying live data sources.

This Claude Skill generates clear commit messages by analyzing git diffs from staged changes. It provides concise summaries under 50 characters with detailed descriptions and affected components. Use it when writing commit messages or reviewing staged changes to follow best practices like using present tense and explaining the "what" and "why."

The Code Reviewer skill analyzes code for best practices and potential issues using Read, Grep, and Glob tools. It automatically checks code organization, error handling, performance, security, and test coverage. Use this skill when reviewing code, checking pull requests, or analyzing overall code quality.

This skill autonomously analyzes uncommitted git changes to decide if and when to commit based on WSP 15 MPS scoring. It generates semantic commit messages that accurately reflect code changes and is triggered by periodic system checks. The skill uses Qwen for strategic analysis and Gemma for validation to ensure pattern fidelity.

This skill provides strategic planning for YouTube telemetry cleanup by analyzing retention classifications from the Gemma agent. It determines optimal cleanup strategies and coordinates safe execution workflows for downstream agents. Use it when you need automated decision-making for telemetry lifecycle management within autonomous operations.

This skill provides fast binary classification of YouTube telemetry records to determine retention strategy. It uses pattern matching to scan heartbeat data as the first phase in a cleanup workflow. Developers should use it for quick initial classification before passing records to downstream agents for retention execution.

This skill audits code, documentation, and planning content for compliance with the WSP framework using a 32K context window. It detects violations and generates corrective guidance, making it ideal for strategic compliance checks. It is designed for production use with MCP orchestration and requires the `pattern_memory` and `libido_monitor` dependencies.

This skill coordinates PQN research by generating hypotheses and synthesizing cross-validation results using Qwen's 32K context window. It processes Gemma PQN emergence detection data and is designed for autonomous research operations during execution phase 3. Use this skill when you need strategic research coordination with pattern fidelity monitoring and MCP orchestration capabilities.

This skill enables Qwen agents to integrate Google research (Scholar, Quantum AI, Gemini, TTS) with local PQN findings for comprehensive validation. It's designed for autonomous operations requiring synthesis of external Google research with internal pattern data. Use this skill when you need to validate findings against Google's research ecosystem while maintaining pattern fidelity.

This skill provides fast binary classification to detect PQN emergence patterns in text, including 0→o artifacts and resonance signatures. It's designed for autonomous operations with a high pattern fidelity threshold of 0.90 and integrates with MCP orchestration. Developers should use it when they need to identify potential emergent phenomena before passing analysis to downstream skills like the qwen_pqn_research_coordinator.

This skill processes high-volume PQN detection data, efficiently handling over 400 raw detections in JSONL format. It performs data summarization and filtering using pattern memory and libido monitor dependencies during autonomous operations. Use this skill in execution phase 4 when you need to prepare processed PQN data for the downstream Qwen research coordinator.

This Claude Skill uses Gemma's pattern matching to detect nested module anti-patterns in filesystems for autonomous monitoring. It performs fast binary classification (<100ms) to identify WSP 3 Module Organization violations when triggered by AI_overseer. Use this skill for real-time detection of nested module patterns during autonomous operations.

This skill coordinates Holo output formatting and telemetry distribution for multiple AI agents. It ensures 0102, Qwen, and Gemma receive properly scoped responses by detecting query intent and managing output modes. Developers should use it when implementing multi-agent systems that require standardized output formatting and telemetry capture.

This prototype skill moderates YouTube live chat by detecting spam, toxic content, and enforcing rate limits. It's designed for testing pattern fidelity with Claude Code before deploying to native Qwen/Gemma environments. Use this to validate the moderation pattern against a 90% success threshold.

This Claude Skill generates automated content for YouTube Live streams, including announcements, chat responses, and moderation messages. It is designed for creating engagement prompts and handling emergency protocol responses during a live broadcast. The skill integrates with systems like live chat and auto-moderation to support stream management.

This skill enables autonomous monitoring of YouTube daemon output to detect Unicode rendering issues like `UnicodeEncodeError` and unrendered codes. It automatically applies fixes using WRE recursive improvement, announces resolutions via UnDaoDu livechat, and triggers daemon restarts. Use this when you need self-healing capabilities for Unicode errors or recursive system improvements in daemon operations.

This skill enhances WSP protocols by using Qwen for strategic analysis and 0102 supervision. It helps developers analyze protocol gaps, generate recommendations, and coordinate multi-WSP updates. Use it when working on WSP enhancements, relying on dependencies like holo_index and wsp_framework.

This skill enables the Qwen agent to mine domain-specific training examples from the 012.txt file, focusing on MPS scoring, WSP patterns, and decision rationale. It operates as part of an MCP orchestration workflow with a high pattern fidelity threshold of 0.90. Use this prototype for autonomous data extraction to prepare training data for the subsequent Gemma domain trainer skill.

This skill audits roadmap files to check completion status, identify missing MCP integrations, and flag outdated skills references. It's designed for autonomous operations and orchestrates MCP tools during the audit process. Use this prototype when you need to validate roadmap integrity before updates.

This skill refactors monolithic CLI applications by extracting command modules from large main() functions using Qwen 1.5B for strategic analysis. It's designed for CLI files exceeding 1,000 lines and reduces main() size by over 70% while preserving functionality. The process includes Gemma validation to ensure pattern fidelity above 90%.

This skill provides strategic cleanup planning for autonomous operations using WSP 15 MPS scoring with WSP 83/64 compliance. It processes labeled files from Gemma's noise detection to generate cleanup strategies during execution phase 2. Key features include pattern memory integration, breadcrumb logging, and a 0.90 pattern fidelity threshold for decision-making.

This skill provides fast binary classification to identify noisy data files (JSONL, logs, temp files) versus useful signal data. It uses pattern matching with a 0.90 fidelity threshold and integrates with MCP orchestration for autonomous operations. Use this prototype when you need to quickly filter out rotting or temporary data files during processing pipelines.