Skill UI

Detects error propagation and chain failures in agent workflows. Identifies vulnerabilities like single-point breakdowns, unbounded retries, and unsafe data propagation to ensure system stability and safety during multi-step operations.

Condense a long agent conversation at a verified phase boundary by writing a durable checkpoint that preserves constraints, evidence, decisions, artifacts and the next action, so a fresh context can resume the same task without repeating work or losing state.

This skill covers splitting extracted text into chunks suitable for LLM context windows or RAG ingestion. It includes configuring chunk size, overlap, and selecting chunker types like markdown, yaml, or semantic. Users can use the standalone chunk command or enable inline chunking during extraction. It supports token-based sizing for specific models and provides configuration options via CLI, config files, or Python API.

Clarity Gate is a pre-ingestion verification system designed for Retrieval-Augmented Generation (RAG) workflows. It ensures documents entering knowledge bases are properly qualified by checking for uncertainty markers and distinguishing assumptions from facts. This skill produces Clarity-Gated Documents (CGD) and validates Source of Truth (SOT) files. It helps prevent hallucinations by enforcing epistemic quality standards before data ingestion. Note that it verifies document form rather than factual truth, requiring human-in-the-loop validation for accuracy.

Use Clarvia AEO (Agent Experience Optimization) to score any MCP server, API, or CLI tool for agent-readiness. It evaluates 15,400+ AI tools across four dimensions—API accessibility, data structuring, agent compatibility, and trust signals—ensuring the components are robust enough for advanced AI agent workflows and dynamic selection processes.

This skill serves as a health assistant for medical information analysis, symptom tracking, and wellness guidance. It helps users manage health queries with structured patterns based on medical data. Users should note this tool is not a substitute for professional medical advice or expert review. It is intended for informational purposes only.

Runs the Hermes Agent CLI in one-shot mode (hermes -z) to delegate bounded local coding tasks from Code Mate. Covers availability checks, timeouts, exit codes, authentication limits and YOLO tool-approval safety constraints.

SOTA computer vision expert specializing in YOLO26, SAM 3, vision language models, and real-time spatial analysis. Provides guidance on designing, implementing, and optimizing advanced vision pipelines, including detection, segmentation, visual reasoning, depth estimation, and 3D reconstruction.

This skill covers strategies for compressing conversation history in AI agent sessions to manage context window limits. It focuses on optimizing tokens-per-task rather than just tokens-per-request. Techniques include anchored iterative summarization, opaque compression, and regenerative summaries. It also addresses artifact trail integrity and evaluation methods like probe-based testing to ensure critical information like file paths and decisions are preserved during compression.

This skill covers predictable performance degradation patterns in language models as context length increases, including lost-in-middle effects, context poisoning, and distraction. It helps diagnose agent failures and design resilient systems by understanding attention mechanics and context engineering strategies for production environments.

This skill covers the core concepts of context engineering for language models, including system prompts, tool definitions, retrieved documents, message history, and tool outputs. It explains attention mechanisms, context windows, and strategies like progressive disclosure to optimize token usage and agent performance.

An intelligent context management specialist for capturing, serializing, and retrieving comprehensive project context across AI workflows. Supports semantic extraction, vector database integration, knowledge graphs, and multi-session collaboration.