Skill UI

Automates Anthropic Admin operations through Composio’s toolkit via Rube MCP, guiding discovery of current tool schemas, verifying connections, and executing schema-compliant workflows with session reuse and pagination awareness.

Automates Anthropic Admin operations through Composio’s toolkit via Rube MCP, guiding discovery of current tool schemas, verifying connections, and executing schema-compliant workflows with session reuse and pagination awareness.

Clarity Gate ensures documents entering RAG knowledge bases retain epistemic quality by generating Clarity-Gated Documents, validating Source of Truth tables, and running deterministic claim/document hash scripts before ingestion.

Formal evaluation framework for Claude Code sessions that applies evaluation-driven development to define capability and regression checks, run deterministic and model graders, and track pass@k/pass^k reliability metrics before shipping.

TransformerLens is the standard library for performing mechanistic interpretability research on large language models. It provides clean interfaces to inspect and manipulate transformer internals—such as attention patterns, residual streams, and MLP outputs—via activation caching and HookPoints. This tool is essential for reverse-engineering model algorithms, performing causal tracing, and studying the internal circuits responsible for model behavior.

This skill optimizes the performance of TwinMind's meeting AI operations. It fine-tunes transcription accuracy and speed using the advanced Ear-3 model, managing audio quality, and implementing caching strategies. Ideal for developers or system administrators needing to integrate, troubleshoot, or optimize high-volume meeting transcription workflows.

A comprehensive, multi-stage workflow for enhancing and structuring raw footage using advanced AI tools. It covers the entire pipeline—from initial recording (Screen Studio) through AI planning (Claude/Codex), deterministic cutting (FFmpeg), compositing (Remotion), asset generation (ElevenLabs/fal.ai), and final human polish (Descript/CapCut). Designed for users creating vlogs, tutorials, or polished corporate videos.

A foresight agent that runs a rigorous 12-step pipeline combining Claude’s qualitative intelligence with Python’s deterministic arithmetic to analyze high-stakes future scenarios, infer reasonable horizons, and surface quantitative signals for prediction requests.

Phoenix Evals lets teams build evaluators for AI/LLM apps by wiring deterministic code with LLM nuance, running Python or TypeScript workloads, validating against humans, and iterating through error analysis, RAG, and production workflows.

A structured methodology for precisely identifying specific geographic locations or historical landmarks. This technique systematically narrows the search scope, moving from broad regions (countries/former states) down through administrative divisions, city centers, and finally to specific monuments or structures. It requires utilizing diverse clues like historical events, census data, and distance constraints, while critically verifying assumptions at each step to ensure accuracy.

A robust orchestration skill designed for managing large-scale, complex machine learning experiments on remote GPU clusters via SSH. It handles multi-seed sweeps, sequential wave transitions, and deep learning pipelines requiring state management. Features include OOM-aware automatic retries, cleanup of stale terminal sessions, and deterministic scheduling to prevent race conditions, making it ideal for grid computing and staged training.

A comprehensive, multi-stage workflow for enhancing and structuring raw footage using advanced AI tools. It covers the entire pipeline—from initial recording (Screen Studio) through AI planning (Claude/Codex), deterministic cutting (FFmpeg), compositing (Remotion), asset generation (ElevenLabs/fal.ai), and final human polish (Descript/CapCut). Designed for users creating vlogs, tutorials, or polished corporate videos.