Skills, prompts, AGENTS.md files and MCP servers that Codex reads. Skills are shared with Claude Code; the rest is Codex-native. 274,431 indexed, measured the same four ways as everything else here.
Expert guide for setting up Claude Code as a complete solo development environment. Covers MCPs, skills, agents, production stack selection, and day-to-day workflows for non-technical product builders who want to ship web and mobile products alone.
Database object patterns for views, stored procedures, functions, triggers, transactions, and dynamic SQL. Use this skill when designing database logic, managing data integrity with transactions, building reusable SQL components, or implementing audit and automation patterns. Covers views, materialized views, stored…
This skill should be used when the conversation is getting long, context feels bloated, Claude mentions compaction or context limits, the user asks about "context", "pruning", "memory", "compaction", or when Claude notices tool outputs are very large. Also activate when resuming a session, starting a large task, or…
Drive an async Claude Code chain — investigate the next slice, run confirm-understanding at a configurable threshold (default 95%), auto-fire via RemoteTrigger if confidence clears, otherwise ping the user. Use when the user has explicitly authorized auto-fire mode (Pattern B+) and asks for the next slice in a chain.
Browser automation CLI for AI agents. Use when you need to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, or testing web apps.
Gives the agent read and write access to a set of allowed directories on the local filesystem. Runs locally from the @modelcontextprotocol/server-filesystem npm package.
Always use when user asks to create, generate, draw, or design a diagram, flowchart, architecture diagram, ER diagram, sequence diagram, class diagram, network diagram, mockup, wireframe, or UI sketch, or mentions draw.io, drawio, drawoi, .drawio files, or diagram export to PNG/SVG/PDF.
Use codex/codex-reply as a persistent second-engineer thread for non-trivial uncertainty — design choices, risky implementation decisions, debugging when stuck, review, spec ambiguity, needing a current/authoritative websearch, or any point where you'd otherwise guess. Keep one thread and iterate; don't treat the…
Scans the current project and generates or updates an AGENTS.md file containing directory structure, build/test/lint commands, coding conventions, and architecture notes. Use when the user says "/init", "initialize this project", "set up project memory", or asks the agent to learn the codebase's conventions for the…
Use when the user wants to practice a U.S. nonimmigrant visa interview as a Chinese citizen, roleplay a consular officer, estimate B1/B2 approval odds, identify 214(b) or 221(g) risk factors, or get concrete interview-prep feedback before a U.S. visa appointment.
A combined safety-analysis method covering vehicle functional safety, limits of intended functionality, and cybersecurity. It maps risks from system failures, difficult operating conditions, and attacks into one analysis.
Plan and implement non-trivial work with a fixed workflow - explore, plan, confirm, implement, validate, review, update docs. Use after write-spec, for direct implementation tasks, bug fixes, or refactors that need a plan.
Use when configuring Claude Code statusLine on Windows with PowerShell, especially when statusLine shows blank/empty or causes terminal layout corruption on exit.
A guide for using Offer岛, a service that collects artificial-intelligence job listings, company information, interview advice, and industry news. It describes the service's API, which is a way for software to request this data.
Write graph-based skills as Durable Workflow Graphs. Use when creating a graph-based skill, converting a linear multi-step skill, or repairing evidence routing, cold resume, or premature completion.
A bridge between a personal WeChat account and Claude Code, so conversations can happen through WeChat. It supports text, images, permission approvals, and slash commands on macOS.
Compile a natural-language spec into a tiny neural function that runs locally with ProgramAsWeights (PAW). Use it for fuzzy text-in / text-out tasks that a regex can't handle but that are too slow, costly, or overkill to send to a full LLM on every item - classify, categorize, label, or tag text (sentiment, urgency…
Spec-Driven Development protocol that runs before every new project or major feature. Creates structured, LLM-optimized specifications that enable one-shot implementation. Guides through 4 sequential phases: Specify, Plan, Tasks, Implement. Detects when specs are needed vs. when to just build. After spec creation…
★not rated 2 2mo agoA237 tokens
originalMIT
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: