trotsky1997

Comprehensive guide for using Claude Agent SDK to build AI agents that can read files, execute commands, edit code, and perform complex workflows. Use when (1) Building autonomous AI agents with Claude, (2) Creating agents that interact with codebases and file systems, (3) Integrating custom tools and MCP servers with Claude, (4) Managing conversation sessions and context with Claude, (5) Implementing file operations, command execution, and code editing in agents, (6) Setting up permission controls and security for agent tools, (7) Using file checkpointing and session management features, (8) Building production-ready agents with error handling and monitoring

Guides developers through using cursor-agent CLI for interactive development, session management, and implementing development-verification loops. Use when working with cursor-agent, creating projects with agent supervision, managing chat sessions, or implementing agent-based development workflows.

Comprehensive DeepResearch methodology for conducting rigorous, traceable research projects with quality gates, structured analysis, and decision-ready deliverables. Use when (1) Conducting deep research projects requiring evidence-based analysis, (2) Managing research progress with quality gates and artifacts, (3) Producing research reports with traceable sources and structured reasoning, (4) Applying OSINT verification techniques, (5) Using structured analytic techniques (ACH, Key Assumptions Check, Red Team), (6) Expressing uncertainty and confidence in research findings, (7) Ensuring research deliverables meet intelligence tradecraft standards (ICD 203/206/208)

Systematic methodology for reading and understanding large codebases efficiently. Use when (1) Understanding a new or unfamiliar codebase quickly, (2) Preparing to modify or extend existing code safely, (3) Debugging complex issues requiring deep code understanding, (4) Onboarding new team members to a codebase, (5) Performing code audits or security reviews, (6) Refactoring legacy code with confidence, (7) Creating documentation for existing systems, (8) Tracing execution flows and data transformations

Guide for creating effective AGENTS.md files for Cursor. Use when (1) Creating new AGENTS.md files in project root, (2) Improving existing agent instructions, (3) Setting up project-specific coding guidelines, (4) Defining code style and architecture rules, (5) Writing simple agent instruction files as alternative to .cursor/rules, or any task related to Cursor AGENTS.md configuration.

Browser automation in Cursor IDE using MCP protocol server `cursor-ide-browser`. Use when (1) Automating web interactions in Cursor IDE, (2) Navigating web pages, (3) Clicking buttons or links, (4) Filling forms or input fields, (5) Taking screenshots or capturing page snapshots, (6) Debugging web pages by checking console messages or network requests, (7) Extracting information from web pages, (8) Testing web applications, (9) Interacting with web-based documentation or tools, (10) Any task requiring programmatic browser control within Cursor IDE

Convert documentation into structured Agent Skills following progressive disclosure principles. Use when (1) Converting user documentation or guides into skills, (2) Creating skills from existing technical documentation, (3) Organizing knowledge into skill format with references, (4) Applying skill design principles to documentation, (5) Structuring information for AI agent consumption, (6) Creating skills that follow best practices for context efficiency

Comprehensive guide for using GitHub CLI (gh) to interact with GitHub from the command line. Use when (1) Managing GitHub repositories, issues, pull requests, or releases from terminal, (2) Automating GitHub workflows in scripts, (3) Creating or managing pull requests, (4) Working with GitHub issues, (5) Creating releases or managing repository operations, (6) Integrating GitHub operations into development workflows, (7) Using GitHub CLI commands in automation scripts

Use CiteClaw to build, clean, expand, and verify bibliographies for surveys, papers, and literature reviews. Trigger this whenever the user asks to collect references, grow a bibliography to dozens or hundreds of papers, clean BibTeX, map literature by topic, turn arXiv/DOI/URLs into citations, or create a survey-style literature map. Use this especially for research writing workflows where CiteClaw alone is not enough and you need a robust process that combines CiteClaw, web search, and manual verification.

Steps that MUST be performed to initialize a Zotero translator script. LLMs cannot generate a valid Zotero translator on their own.

Steps that MUST be performed to verify a Zotero translator after every addition or modification, and before submitting a PR.

Run an autonomous optimization loop on Claude Code. Frames a need into a measurable target, then loops (hypothesis → measure → keep/discard) with no human in the loop until the goal is met. Use when asked to "optimize X in a loop", "run autoresearch", or "experiment until Y improves".

Use when an autonomous coding agent (no human in the loop) needs to drive OpenSpec spec-driven development end-to-end — bootstrap, propose, implement, and archive a change without slash commands or interactive prompts. Triggers on "用 openspec 自主开发", "agent-own spec-driven dev", "openspec 无人值守 / 非交互", "spec-driven 自动跑一遍", "openspec init/propose/apply/archive 全自动", "drive OpenSpec from an agent". Captures the verified non-interactive CLI loop, the status/instructions state machine, the substitutions for OpenSpec's built-in "ask the user" stops, and validation-driven self-correction.

Run an autoresearch-style loop where an optim agent tunes an HCP worker harness for one task metric by changing agents_md, worker hooks, and the skills portfolio. Use when asked to tune an HCP worker agent, improve task pass rate, optimize harness behavior, or run an HCP optimization loop.

Configure `lsp-pi` for a project or user environment. Use this whenever the user wants help setting up Pi LSP, formatters, analyzers, Python provider selection, or `.pi/settings.json` / `~/.pi/agent/settings.json` entries for code intelligence. Also use it when the user asks which `lsp-pi` servers, formatters, or analyzers to enable for a repo.

Use the project Wolfram CLI for Wolfram Alpha style lookups, symbolic math, equation solving, ODEs, integrals, derivatives, unit conversions, and result inspection. Make sure to use this skill whenever the user asks for Wolfram Alpha behavior, math-heavy reasoning, conversions, or factual lookups that fit WA better than ad hoc web or manual HTTP requests. Prefer the CLI commands over hand-built API calls because they handle encoding, result extraction, and project-local persistence.

Use when the user asks to fetch a web page as Markdown, cache web content locally, crawl a bounded documentation subtree, mirror docs under .md/, or use the trotsky1997/webfetch-cli GitHub npx tool. This skill is for web-to-Markdown retrieval without Pi.

This skill should be used when code search is needed (whether explicitly requested or as part of completing a task), when indexing the codebase after changes, or when the user asks about ccc, cocoindex-code, or the codebase index. Trigger phrases include 'search the codebase', 'find code related to', 'update the index', 'ccc', 'cocoindex-code'.

Interview the user to design a project-level Hookify V2 or Pi hook rule set for a repository. Use this whenever the user wants help defining `.pi/hookify.*.local.yaml` rules, choosing `on / when / do / respond`, designing repo-wide guardrails, approvals, command or file safety checks, prompt transforms, rollout policy, or asks things like "design my hook rules", "what hooks should I enable", "repo guardrails", "制定项目级 hook 规则", or "帮我设计项目级 guardrails", even if they do not mention Hookify by name.

Turn a repeatable workflow from the current conversation into a reusable standard skill. Use when the user wants to capture what you just did as a skill, turn a successful session into a `SKILL.md`, preserve a reusable process, or says things like "make this a skill", "turn this into a skill", "capture this workflow", or "productize this process."

Interactive consultant for adopting spec-driven development in a repository. Use when the user is unsure how to start with specs, needs interview-style requirements discovery, wants help choosing the right `/spex-*` or `/speckit.*` command, or asks how to apply spec-driven development to a new or existing project.