Shape agent capabilities into your workflow.

Econ Writing Skill turns advice from more than 50 economics writing guides into an agent skill for Claude Code and Codex. It can draft paper sections, rewrite vague or passive prose, review structure and style, and mark places where the author must supply actual results. It is most useful when the research content already exists and the task is writing, revision, or audit.

The repository currently provides the `codex-dynamic-workflows` skill for supervised AI-agent workflows. It can organize a goal into subagents or simulated work packets, pause at approval gates, integrate results in the parent flow, verify the outcome, and retain reusable workflow artifacts. Users can install the skill in a compatible agent and invoke it when a task benefits from staged coordination and explicit oversight.

h5i hooks into Claude Code or Codex to version human prompts, agent context steps, and trimmed tool output as recoverable Git objects. It can also place several agents in isolated worktrees, coordinate peer review, test competing implementations, and apply the selected result. The workspace evidence lives inside repository refs, so it can be pushed, fetched, inspected offline, and used without depending on a hosted development environment.

She Love Me is an agent skill that extracts WeChat or QQ conversations and applies frameworks such as attachment styles, Gottman, and Sternberg to produce relationship-pattern analysis, risk signals, and practical suggestions. The results are guidance, not a definitive judgment. Its agent workflow emphasizes local processing; the traditional path exports messages.json and an analysis prompt, and whether those files leave the device depends on the chat model selected. Windows supports both platforms, macOS WeChat support is experimental, and Linux currently relies mainly on QQ.

The editorial-card-screenshot skill reorganizes source material into editable, screenshot-ready HTML cards and can optionally render PNG output. It supports portrait, landscape, square, widescreen, and banner-like aspect ratios by adapting the information layout rather than merely resizing one template. Installation requires the complete skill directory, including its assets, references, scripts, and optional agent files.

SkillAnything is a meta-skill that turns a target such as a CLI tool, REST API, package, workflow, or web service into agent-skill packages. Its seven-phase process analyzes the target, designs the skill architecture, implements SKILL.md and supporting files, creates tests, benchmarks the result, optimizes triggering, and packages distributions. The output can target Claude Code, OpenClaw, Codex, and a generic skill format, reducing the manual work of adapting the same software capability across agent platforms.

OK Skills packages 32 reusable skills plus AGENTS.md and CLAUDE.md playbooks for Codex, Claude Code, Cursor, and other compatible tools. The library covers documentation lookup, browser automation, planning, debugging, frontend work, and office documents, while some skills still depend on tools such as gh, a browser environment, or external MCP services.

Repo Analyzer scans a project, asks the user to choose an analysis depth, and can dispatch parallel subagents across core modules. It emphasizes why the architecture was designed a certain way, compares trade-offs with alternatives, and saves a single Markdown report with Mermaid diagrams. External search MCP servers are optional rather than required.

Web Asset Generator creates favicons, app icons, and social sharing images from a logo, text, or emoji, then produces common sizes and checks file and accessibility requirements. It can detect frameworks such as Next.js or Astro and help integrate the assets. The scripts require Python 3.6+ and Pillow; emoji rendering uses optional dependencies.

AgentSilex is a roughly 300-line Python framework built around readable Agent, Runner, Session, and tool abstractions. LiteLLM lets it switch among more than 100 model providers, while the framework adds tool calls, session history, multi-agent handoffs, MCP clients, streaming, and structured output. OpenTelemetry tracing is included for inspecting execution, and the package requires Python 3.12 or newer.

Compose Performance Skills teaches a coding agent how to measure, diagnose, fix, and verify Jetpack Compose performance problems. The library covers stability, recomposition, lazy layouts, custom modifiers, side effects, baseline profiles, R8, and related tooling, with references back to primary sources and version-pinned APIs. It is aimed at Android Compose rather than general onboarding, and Compose Multiplatform targets have not been specifically validated.

agr treats agent skills like project dependencies for teams. It installs skills from Git repositories or local directories into detected AI tools, records the desired set in `agr.toml`, and pins exact versions in `agr.lock` so teammates can reproduce the same environment with one sync command. Users can also upgrade skills deliberately or run one temporarily with `agrx` without adding it to the project configuration.

OpenSkills exposes Claude-style skills through an MCP server and executes them in an isolated local container, allowing assistants such as Claude Desktop, Gemini CLI, or Qwen CLI to process documents and images without cloud uploads. Path mapping lets existing skills run without modification and returns files to a local outputs directory. It currently requires an Apple Silicon Mac, macOS, and Python 3.10 or newer.

OpenAgents is developing Sarah as an inspectable supervision layer for bounded plans distributed across named Codex, Claude, and Grok accounts, using local desktops or optional cloud capacity. The repository is early and changing quickly, and Sarah does not yet compose the existing fleet components into the complete daily workflow. Its current milestone is to manage at least three useful work streams at once.

Swift Testing Agent Skill gives compatible AI coding tools focused guidance for Apple's modern Swift Testing framework. Its references cover test structure, expectations, asynchronous workflows, parameterized tests, traits and tags, parallel execution, isolation, and incremental migration from XCTest. Teams can use it during implementation or code review to identify migration opportunities and flaky patterns. It installs through the Agent Skills CLI, a Claude Code plugin, or manual copying, and functions as guidance rather than a standalone test runner.

Rampstack’s library connects brand discovery, information architecture, content, design, frontend delivery, launch operations, analytics, and growth into a consistent website lifecycle. Its 103 skills share common conventions and are designed to work across stacks such as Next.js, WordPress, Shopify, Webflow, or plain HTML. Most guidance is tool-agnostic, while the dedicated SEO audit suite expects an Ahrefs MCP connection and may consume paid API credits during long agent sessions.

CSV Data Summarizer is a Claude Skill that inspects an uploaded CSV and produces descriptive statistics, correlations, distributions, trends, and missing-value checks without first requiring an analysis prompt. It selects charts from the available date, numeric, and categorical columns, including time-series plots, heatmaps, and histograms. Claude users install the release ZIP as a Skill, while running the included analyzer directly requires Python and its listed dependencies.

Add a modular set of Claude skills for tasks such as SEO reporting, sitemap creation, document parsing, Docker setup, resume or report generation, and CI/CD configuration. The npm package attempts to copy skills into the host project during installation and includes scripts for rebuilding package and index files.