Skill UI

Transforms Claude Code into a continuous autonomous agent with persistent memory, scheduled operations, remote dispatch, computer automation, and task queuing for repeatable workflows and monitoring.

A standardized, multi-phase workflow designed for initiating large-scale, complex software tasks (e.g., migration, overhaul, refactoring). This process ensures that all necessary planning, project analysis, risk assessment, and task decomposition are completed *before* any actual coding begins. It guides users through intent confirmation, deep codebase analysis, and creating a detailed, trackable development plan.

Workspace Surface Audit scans repo files, MCP servers, plugins, connectors, and environment configs to report current capabilities and recommend ECC-native skills, hooks, agents, or workflows for setups like Claude Code.

Accelerates CPU-bound Python workflows by translating NumPy, pandas, ML, graph, image, simulation, and IO code into NVIDIA GPU kernels using CuPy, Numba CUDA, Warp, cuDF/cuML/cuGraph, KvikIO, cuSpatial, and RAFT for dramatic speedups.

Extracts business domain knowledge from a codebase, optionally reusing an existing knowledge graph, and generates an interactive horizontal flow graph for dashboard visualization, with lightweight scanning and validation steps.

Turns your existing Kanban board into an autonomous AI-driven development pipeline, coordinating Claude Code workers, deterministic gates, cost tracking, and crash-proof orchestration for full lifecycle work.

Guides creation of persona-targeted VS Code CodeTour .tour files, covering repo exploration, intent inference, step composition, and validation so each narrative walkthrough is anchored to real files and tailored depth.

Guides ECC agents through live message retrieval, inspection, and code recovery workflows while ensuring the exact messaging surface, thread, and status are documented before replying or proving a source.

Keeps structured JSON records for every code change, enforces a state-machine workflow, and hands off AI sessions with summaries, blockers, and files so bots can resume seamlessly after interruptions.

Guides applying Dart 3 branches, patterns, pattern types, and records when writing Flutter/Dart code, covering if-case, switch expressions, guards, and record usage to encourage correct, exhaustive handling in control-flow and destructuring scenarios.

Guidelines for Go code clarity covering line breaks, variable declarations, control flow, function design, and composite literals to keep code consistent during development or reviews.

This skill provides robust mechanisms to verify the integrity of AI agent plugins, tools, and dependencies. It generates deterministic SHA-256 manifests (INTEGRITY.json) for all components, allowing users to detect unauthorized modifications, track provenance, and ensure that deployed assets match their published versions. Use this during CI/CD pipelines, code review, and pre-production promotion to secure the entire agent ecosystem.