Skill UI

Utilizes structured critical reasoning to rigorously stress-test ideas, plans, and decisions. This skill allows you to play devil's advocate, conduct pre-mortems, execute red-teaming exercises, or audit underlying assumptions. It employs multiple advanced modes, including Socratic questioning, Hegelian dialectics, and Falsificationism, to identify blind spots, weaknesses, and unstated assumptions before commitment.

Implements middleware policies to throttle API requests, limit concurrency, queue priorities, and use circuit breakers with adaptive limits, protecting backends from overload while surfacing throttle-state headers for clients.

Automate TikTok content tasks via the Rube MCP-powered Composio toolkit: upload and publish videos, post photos, monitor publish status, list your media, and pull profile/stats data while following required sequencing and authentication steps.

Provides a structured reference covering the 100 most critical web vulnerabilities, including definitions, root causes, impacts, and mitigation guidance aligned with OWASP categories to support assessments and remediation planning.

torchforge is Meta's advanced PyTorch-native library designed for agentic Reinforcement Learning (RL). It fundamentally separates core RL algorithms from complex distributed infrastructure concerns. This allows researchers to rapidly prototype and experiment with novel RL methods (like GRPO) using clean, PyTorch-native abstractions, while automatically handling large-scale tasks like distributed training across thousands of GPUs using Monarch and TorchTitan.

Tracks and optimizes application response times using the response-time-tracker plugin, covering API endpoints, database queries, external calls, frontend rendering, and background jobs to highlight bottlenecks or SLO breaches.

Tracks CPU, memory, disk/network I/O, DB connections, and threads via the resource-usage-tracker plugin to reveal performance bottlenecks, recommend right-sizing, and guide cost-saving actions.

Automates the creation of technical troubleshooting guides by following documentation best practices, generating templates, and validating outputs for API docs, user guides, and architecture references whenever you mention troubleshooting patterns.

Guides Claude to automatically tune machine learning model hyperparameters via grid, random, or Bayesian search, generating runnable Python code, executing validations, and reporting best configurations plus performance metrics for accuracy or other goals.

Configure Turborepo caching for monorepo projects with templates for global and package pipelines, remote cache usage, self-hosted servers, and scoped filters to speed CI/CD builds.

Automates Turbot Pipes tasks by routing commands through Composio's Rube MCP toolkit, guiding you to discover current tool schemas, verify Turbot connections, and execute multi-step workflows safely.

Guides tutorial engineers through pedagogical design, concept decomposition, and exercise scaffolding so codebases turn into hands-on learning journeys with checkpoints, examples, and troubleshooting.