Progressive-disclosure instructions an agent loads on demand. Cheap until invoked, which is why the per-session figure on each card is usually two digits.
Skill Codex
Use the bundled native CLI over HTTP REST to retrieve and preserve Manifold memory. Read this skill before checking availability; Manifold has no MCP tools or gRPC transport. Use before non-trivial work for prior decisions, procedures, incidents, and evidence, and after verified work for durable knowledge. Supports…
Skill Codex
Interactive bilingual course skill for learning loop engineering through Loopath. Use when a user wants to start, continue, install, teach, study, verify, or be quizzed on the Loopath course; when they ask for an agent-guided course on coding-agent loops, harness design, AGENTS.md, action schemas, tool use, policy…
CCDevelopForFun/agent-controller
Skill Claude CodeCodex
Teaches the agent to format all times as ISO-8601 UTC with an explicit timezone label.
phenomenoner/hermes-agent-harness-plus
Skill Claude CodeCodex
Use for compaction, coordination, or two complexity cues.
wangling-miao/codex-usage-receipt
Skill Claude CodeCodex
Use this skill whenever the user asks for Codex token usage, Codex cost, token consumption since a date, remaining Codex quota/limits, cache-hit tokens, model-by-model pricing, or a receipt/report/PDF/printout for Codex usage. This skill should also trigger for Chinese prompts asking to print a Codex token spending…
Skill Codex
Run, supervise, or explicitly cancel long-running, background, async, or detached processes from OpenAI Codex CLI or the Codex VS Code extension and wake or resume the owning Codex session on process completion, without LLM/model polling while the job runs. Use when asked to run something in the background, start or…
Skill Codex
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Codex's capabilities with specialized knowledge, workflows, or tool integrations.
hou361996193-cloud/wechat-codex-bridge
Skill Claude CodeCodex
Use when setting up, starting, stopping, checking, or troubleshooting the local WeChat bridge for Codex CLI.
Skill Claude CodeCodex
Use when the user wants Codex to split work across multiple subagents, especially for implementation plus exploration or verification in parallel.
Skill Claude Code
Shows you something worth reading while your coding agent works, then notifies you when it's done. Use this skill to install turnbreak's turn-start/turn-stop hooks for Claude Code, Codex, Gemini CLI, or Copilot CLI, or to manage its interests and settings.
uninhibited-scholar/obsidian-loop-wiki
Skill Claude CodeCodex
Ingest and grow an Obsidian knowledge base as a verifiable loop. Combines Loop Engineering (machine-checkable completion criteria, boundary guardrails, a verification pass that resists gaming) with the Obsidian "second brain" organizing model (typed nodes, wikilinks, MOC, an index and an audit log). Use this whenever…
Skill Claude CodeCodex needs its repo
Gather deterministic architecture-fitness evidence with archfit: full checks, scorecards, deltas, SARIF/JSON findings, tool coverage, and agenttasks. Use when a repo has .archfit.yaml or the user wants architect to combine its review with archfit. Produces hard facts for architecture-review to calibrate, not a…
Skill Claude CodeCodex
Use when the user wants to plan, dispatch, review or deploy software sprints in a multi-chat orchestration pattern — running parallel multi-agent work across isolated git worktrees without context bloat in long Claude sessions. Skill provides a portable workflow with auto-discovery of project profile, modular addons…
Skill Claude Code
A Fable-5-inspired work-discipline protocol adapted for Codex CLI and Codex desktop. It uses a plan gate, small-card execution, adversarial self-check, real-product verification, context hygiene, and checkpoint autonomy. Activate ONLY on an explicit request — the user names the mode ("use fable mode", "enable…
0xCyberBerserker/token-rat-esp
Skill Codex
Aggressive token and output minimization for coding agents. Use when the user wants concise technical Spanish, minimal useful output, small patches, and strict scope control.
Skill Codex
Use when a task needs the full OpsX workflow from exploration through change/spec, planning, implementation, verification, archive, and branch finish using OpenSpec and superpowers.
TanChuping/codex-task-watchdog
Skill Codex
A monitoring and troubleshooting skill for long-running Codex tasks. Codex is a coding agent that can run tools and background work.
thoreinstein/extension-creator
Skill Claude CodeCodex
Use this skill when the user wants to create, scaffold, or configure a new Gemini CLI extension. It provides expert guidance on manifest structure, MCP servers, Agent Skills, and release workflows.
YoungApple/growth-retrospective-skill
Skill Claude CodeCodex
Personal growth gap retrospective. 5-category sweep ranked by decision-velocity signals; each top gap launched with a Deep Research / NotebookLM prompt + 4 multiple-choice questions grounded in your own codebase (your files + ADRs are the ground truth, not the AI's knowledge). Step 0 Action Audit refuses new items if…
Skill Claude CodeCodex
A voice-control rule for a gated Korean-and-English voice assistant. It defines which wake phrases start listening and how spoken replies must be formatted.
bakhtiersizhaev/chromewire-mcp
Skill Claude CodeCodex
Install and verify ChromeWire MCP, the local bridge between the official Codex Chrome Extension and MCP-compatible AI CLI tools.
Skill Claude CodeCodex
The Chariot VII — The Narrative Breaker. Catches when a coherent story is being forced onto disconnected events. Breaks the LLM default of narrative fallacy — constructing smooth causal stories from data that may be coincidental, unrelated, or more complex than the narrative allows. Use this skill when a post-mortem…
Skill Claude CodeCodex
The definitive Hebrew content skill for AI agents. Write Israeli Hebrew that sounds native — not Academy Hebrew, not translated English. Covers all registers (casual to formal), audience segments (secular, traditional, religious, Haredi, Russian-speaking, Arab-Israeli, Anglo), platform strategies (Instagram-first…
Skill Claude Code
Guidance for running commands that stay active, such as development servers, file watchers, and watch-mode builds, in a separate visible terminal tab. It is written for the oretachi development interface, where users can view logs and stop the process.
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: