Skill UI

This skill enables parallel read-only multi-agent root-cause investigation for bugs, regressions, crashes, or unexplained failures. It deploys four specialized sub-agents to analyze reproduction, code paths, recent changes, and proof plans simultaneously. The main agent synthesizes findings into ranked hypotheses without editing files. Ideal for diagnosing complex software issues efficiently.

Guide for integrating BuyWhere's MCP and API surfaces into AI shopping agents, enabling product search, price comparison, and deal discovery across Cursor, Claude Desktop, or custom runtimes.

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.

This skill transforms completed debugging sessions or repeated corrections into evidence-backed procedures. It captures reusable lessons upon explicit request, ensuring privacy by not automatically saving memories. Users must provide session summaries or transcripts to identify failures and verify fixes. The output is a reviewed procedure suitable for documentation, excluding secrets and personal details. It helps maintain knowledge without modifying agent configuration automatically.

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 enables AI agents to control the user's visible browser interface directly. It supports navigation, clicking, typing, and debugging actions within the browser pane. The agent can take screenshots, interact with web tools, and handle login processes. It operates via a live Automation REPL without requiring code mode, ensuring safe and scoped interactions within the current session.

Score any MCP server, API, or CLI for agent-readiness using Clarvia AEO (Agent Experience Optimization). Evaluate tools across four dimensions: API accessibility, data structuring, agent compatibility, and trust signals. Supports scoring, searching, comparing, and leaderboard queries.

Deep expertise in Anthropic's Claude Code CLI: configure hooks, MCP connections, permissions, CLAUDE.md memory, sub-agents, and advanced workflows to maximize developer productivity.

This skill reviews generated or changed production code against Clean Code principles like SOLID, DRY, and KISS. It checks for LLM-specific failure modes to ensure code quality before shipping. It operates in three modes: guard-pass after generation, live during writing, and review for auditing. It helps AI agents avoid systematic errors and maintain high standards without replacing project linters.

This skill enables orchestrators to delegate bounded coding tasks to the Cline CLI agent while retaining review and landing control. Users write a brief, dispatch via a relay script, wait for completion, review the generated diff, and commit changes manually. It requires the Cline CLI installed and authenticated, ensuring human oversight in the coding workflow.

Control cmux workspaces, panes, surfaces, and agent sessions from macOS terminal workflows, including topology inspection, input routing, notifications, and sidebar metadata.

Run DeepSeek Harness non-interactively to delegate coding tasks to a headless agent. Covers availability checks, timeouts, prompt passing, exit codes, authentication limits, and safe read-only versus write workflows.