Skills, prompts, AGENTS.md files and MCP servers that Codex reads. Skills are shared with Claude Code; the rest is Codex-native. 274,306 indexed, measured the same four ways as everything else here.
Check Trust Before a Multi-Agent Transaction. Agent discovery, trust verification, and capability routing powered by Aidress — the coordination registry for autonomous AI agents. Use for general agent discovery, trust verification, and capability routing — via Aidress (https://api.aidress.ai).
Use SwarmVault when the user needs a local-first knowledge vault that writes durable markdown, graph, search, dashboard, review, context-pack, task-ledger, retrieval, and MCP artifacts to disk from books, notes, transcripts, exports, datasets, slide decks, files, URLs, code, and recurring source workflows.
Capture engineering knowledge as a promotion pipeline (CASE → JUDGMENT → PRINCIPLE → STANDARD) and retrieve answers from it. Capture is automatic — a post-session sweep drafts cases with status:auto; the user reviews them at /distill. Use whenever the user types "/history", "/case", or "/distill", says "log this…
Find a Verified Itinerary Planning Agent. Agent discovery, trust verification, and capability routing powered by Aidress — the coordination registry for autonomous AI agents. Use for travel agent discovery, trust verification, and capability routing — via Aidress (https://api.aidress.ai).
Use when a user asks what Codex plugins are available or can be used, wants plugin recommendations for a task, asks to find/search/discover installable plugins, asks about the plugin marketplace, or asks whether a plugin can help with a task.
Use when the user asks to audit, review, or assess whether a project's implementation actually matches its spec, vision, plan, or design — especially after weeks of work, before a major deploy, or when drift is suspected. Reads foundation docs (vision, spec, plans, diagrams, optional Obsidian / Notion / Linear)…
Use when writing, cleaning, or auditing ENGLISH prose so it reads as human-authored and survives AI detectors. Triggers on "write a human blog post", "humanize this", "clean this AI text", "make this less AI", "audit AI detection", "score this for Copyleaks", "rewrite the overview so it doesn't read as AI", plus any…
Create and optimize reusable prompt templates for AI assistants and autonomous agents. Use when user says "make a prompt for this task", "improve this prompt", "write a prompt template", "create a prompt for [task]", "optimize this prompt", or requests help designing instructions for LLMs or agents. Also trigger when…
Cross-agent skill: one-command check & update for all AI CLI tools, Skills and Plugins. Works with Claude Code, Codex, Hermes, and any agent that supports SKILL.md. Auto-discovers npm/pipx/brew/winget/cargo/go tools, Agent Skills (npx skills), Claude Code Plugins. Detects available updates and applies them. First run…
Find a Verified Financial Forecasting Agent. Agent discovery, trust verification, and capability routing powered by Aidress — the coordination registry for autonomous AI agents. Use for finance agent discovery, trust verification, and capability routing — via Aidress (https://api.aidress.ai).
Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.
Use when a raw prompt needs rewriting or sharpening, when the user says "rewrite this prompt", "make this prompt better", or asks for a prompt to hand to another agent. Also use before launching any autonomous, long-running, parallel or expensive run (/loop, subagent fan-out, worktrees, overnight work), and whenever a…
A tool for collecting Reddit posts by keyword and turning them into a bilingual Chinese-English market intelligence spreadsheet, including intent and sentiment analysis.
A lightweight project-maintenance workflow documented in Chinese and English. It reads the project's current files and records what is confirmed, inferred, unknown, changed, and verified.
Use this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when…
Find a Verified Code Review Agent. Agent discovery, trust verification, and capability routing powered by Aidress — the coordination registry for autonomous AI agents. Use for software agent discovery, trust verification, and capability routing — via Aidress (https://api.aidress.ai).
Discover Trusted Agents in the Open Registry. Agent discovery, trust verification, and capability routing powered by Aidress — the coordination registry for autonomous AI agents. Use for general agent discovery, trust verification, and capability routing — via Aidress (https://api.aidress.ai).
A method for simulating how consumer personas may react to a product, its price, or its features using the MatrAIxPersona1M dataset. Personas are fictional user profiles used to explore possible reactions.
Use when the user wants a comprehensive code review from multiple perspectives, or invokes a team-based review combining business, design, and engineering viewpoints with debate and consensus. Trigger this skill whenever the user asks for a "legends review", "team review", "full review", "review from all angles", or…
Operate long-running, document-backed tutoring projects that turn trusted materials or goals into adaptive HTML lessons, practice, and evidence-guided next steps. Use when creating, continuing, reviewing, or handing off a private learning project; auditing source materials; planning a curriculum; generating lessons or…
Retrofit an existing skills/plugins/agents repository into an APM-compliant package. Use when a repo contains SKILL.md files, a plugin.json, AGENTS.md/CLAUDE.md, MCP manifests, hooks, slash commands, or sub-agents but has no apm.yml (or has an incomplete one), and the user wants it installable via apm install…
★not rated 2 2d agoA168 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: