Skill UI

A sophisticated, two-stage system designed to capture and internalize user corrections and learnings during ongoing conversations. It automatically detects patterns (e.g., "use X instead of Y") and queues them. Users can manually review and apply these learnings to a persistent knowledge base (CLAUDE.md), ensuring the AI continuously improves its knowledge and adherence to specific user context for future sessions.

This skill implements an autonomous, multi-round review loop for refining complex research or codebases. It iteratively reviews content using specified external backends (e.g., Codex, manual review), implements necessary fixes based on feedback, and re-reviews until the positive assessment criteria are met or the maximum round limit is reached. It manages state persistence and allows for configurable difficulty levels, making it ideal for projects requiring rigorous, self-correcting quality assurance.

Applies two-axis scoring, mandatory devil’s advocate reasoning, and persistent history to keep post-task self-reviews honest, resistant to inflation, and suitable for Claude Code sessions.

AgentFlow transforms your existing Kanban board (Asana, GitHub, Linear) into a self-managing, autonomous AI development pipeline. It orchestrates multiple AI agents through defined stages (Backlog to Done), enforcing deterministic quality gates, performing adversarial reviews, and tracking costs—all visible within your existing project management UI. It provides crash-proof execution and full observability for complex software development lifecycles.

This skill activates the P7 Senior Engineer mode, designed for rigorous, solution-driven execution under P8 supervision. It is used when the user explicitly requests 'P7模式' or when invoked as a sub-task executor by P8. The process involves three key stages: designing a detailed implementation plan and impact analysis, executing the core coding, and finally, completing a mandatory three-question self-review, ensuring high quality and comprehensive verification.

A multi-phase workflow designed to guide users through complex problem analysis and solution design. Instead of providing immediate answers, it systematically enforces deep thinking by reviewing the problem's validity, assessing information gaps, performing root cause analysis, and iterating through solution vetting (self-check and final review). Ideal for complex technical troubleshooting, decision-making, or theoretical discussions requiring rigorous methodology.

This skill manages comprehensive identity verification using the Persona API. It covers government ID checks (passport, driver's license), selfie liveness detection, and database lookups (SSN, watchlist). Use this when implementing Know Your Customer (KYC) workflows, reviewing detailed verification results, or building custom approval logic based on multi-source identity data.

This framework enforces high output quality, intellectual honesty, and continuous improvement across all AppSec agents. It mandates structured processes, including self-critique loops, anti-rationalization guards, and self-reflection quality gates (≥8 scoring). Applicable during code review, SAST, SCA, and threat modeling, it ensures rigorous and verifiable security analysis.

This tool assists engineers in transforming vague work logs into structured, evidence-backed impact statements for professional documentation. It uniquely mines Copilot CLI session logs, git commits, and GitHub PRs to reconstruct forgotten work. It enforces a mandatory 'Action → Result → Evidence' impact contract, making it indispensable for preparing performance reviews, promotion packets, and self-assessments.

Generates a comprehensive, single-screen financial report in a self-contained HTML file. This skill is designed for analyzing key business metrics, including P&L summaries, MRR trends, KPI cards (Revenue, Gross Margin, Cash Runway), and top-accounts analysis. Ideal for quarterly or monthly financial reviews in SaaS or startup environments.

Drafts self-assessments, peer reviews, and upward feedback using WorkIQ evidence from emails, meetings, and contributions. It helps you write honest, impact-focused performance reviews in your own voice with STAR-based structure.

This alias serves as a specialized Codex CLI interface for implementing the P7 protocol, mimicking the workflow of a P7 senior engineer. It enforces a highly structured development cycle: detailed planning, robust implementation, mandatory self-review (via three critical questions), and final, verifiable completion markers. Ideal for complex coding tasks requiring maximum rigor and accountability.