Autonomously improve a generated paper via GPT-6-Astra xhigh review → implement fixes → recompile, for 2 rounds. Use when user says "改论文", "improve paper", "论文润色循环", "auto improve", or wants to iteratively polish a generated paper.

Generate a structured paper outline from review conclusions and experiment results. Use when user says "写大纲", "paper outline", "plan the paper", "论文规划", or wants to create a paper plan before writing.

Analyze ARIS usage logs and propose optimizations to SKILL.md files, reviewer prompts, and workflow defaults. Outer-loop harness optimization inspired by Meta-Harness (Lee et al., 2026). Use when user says "优化技能", "meta optimize", "improve skills", "分析使用记录", or wants to optimize ARIS's own harness components based on accumulated experience.

Azure AI Agents Persistent SDK for Java. Low-level SDK for creating and managing AI agents with threads, messages, runs, and tools.

Use when a user asks for a plan for a coding task, to generate a clear, actionable, and atomic checklist.

Send SMS messages with Azure Communication Services SMS Java SDK. Use when implementing SMS notifications, alerts, OTP delivery, bulk messaging, or delivery reports.

Azure AI Agents Persistent SDK for Java. Low-level SDK for creating and managing AI agents with threads, messages, runs, and tools.

Email infrastructure for AI agents. Create accounts, send/receive emails, manage webhooks, and check karma balance via the AgentMail API.

Fan-out search across all memory sources when context is unclear or vaguely referenced. Triggers on: 'from earlier', 'remember when', 'what we discussed', 'that thing with', 'the conversation about', 'did we ever', 'what happened with', 'you mentioned', 'we talked about', 'earlier today', 'last session', 'the other day', or any vague reference to past context that needs resolution before the agent can act.

Get the current public/outbound IP address of this machine using a reliable shell workflow with multiple fallback endpoints. Use when users ask for outbound IP, public IP, WAN IP, egress IP, or need IP checks before firewall/network allowlisting.

Expert documentation generation for golden layers. Detects SCD types, documents business rules, metric definitions, aggregation logic, and data quality scoring. Use when documenting golden layer tables.

Use when working with ClawFreelance - the AI agent freelancing platform. Helps agents discover tasks, find bounties, claim work, submit solutions, and check reputation.

Quick-reference card for caveman modes, skills and commands. Trigger: /caveman-help or "caveman help".

Set up and configure MCP (Model Context Protocol) servers with Claude Code. Use when the user wants to connect Claude Code to external tools, databases, APIs, or services via MCP. Handles HTTP, SSE, and stdio server configurations with proper authentication.

Create a new Claude Code agent with proper YAML frontmatter structure. Use when the user wants to add a specialized agent to a plugin. Handles agent file creation with name, description, tools, model selection, and color configuration.

JPA/Hibernate patterns and common pitfalls (N+1, lazy loading, transactions, queries). Use when user has JPA performance issues, LazyInitializationException, or asks about entity relationships and fetching strategies.

Build typed reactive forms with strict FormGroup typing, custom validators, and nonNullable controls in Angular. Use when implementing typed reactive forms, custom validators, or form control patterns.

Write Android unit, Compose UI, and Hilt-integrated tests. Use when designing test behavior with MockK or coroutine test utilities; defer database/WorkManager-specific recipes to the owning feature skill.

Development tools, linting, and build config for TypeScript. Use when configuring ESLint, Prettier, Jest, Vitest, tsconfig, or any TS build tooling.

Analyze conversation corrections to detect skill gaps and prepare targeted skill-library maintenance tasks. Use after any session with user corrections, rework, or retrospective requests. After finding correction loops, also load +common/common-learning-log to persist mistake entries to AGENTS_LEARNING.md.