AI Agent Skills Marketplace
24 skills and agents available
ai-productivity
The AI Productivity skill acts as a high-performance intake layer for your agentic workflows. It identifies and resolves common execution bottlenecks before they waste tokens or time: vague prompts, overwhelming context "bloat," and high-risk ambiguous requests. Instead of diving blindly into a task, this skill triages the work, compresses relevant data, or rewrites instructions into actionable prompts.
truth-first-lite
The Truth-First Verification skill enforces a rigorous, evidence-based workflow for AI agents. It prevents "hallucinated certainty" by requiring real-time verification of system states, configurations, and file contents before the agent is allowed to provide an answer. Instead of assuming a service is running or a config is correct, the agent must prove it using available diagnostic tools.
Git Commit Assistant
Writes structured, conventional commit messages from git diff output. Supports Conventional Commits, semantic versioning, and project-specific commit styles.
Code Review Agent
Comprehensive code review with security, performance, and style analysis for Python, TypeScript, and Go. Detects OWASP Top 10 vulnerabilities, N+1 queries, race conditions, and logic errors before they reach production.
prompt-engineer-lite
Building high-performance LLM applications requires more than just basic instructions. This skill equips your AI agent with a sophisticated framework for designing, debugging, and optimizing prompts across any major model provider. It solves the common problems of model drift, parsing failures, and hallucination by implementing industry-standard engineering patterns.
pretext-layout
The Pretext Layout skill enables AI agents to integrate and debug high-performance multiline text measurement using the @chenglou/pretext library. It solves the common frontend performance bottleneck of "layout thrashing" by replacing slow DOM-based height probes with a fast, browser-native API approach.
PR Description Generator
Generates comprehensive, structured PR descriptions from git diff or branch comparison. Includes summary, changelog, testing notes, and deployment considerations.
Independent Website SEO & GEO Auditor
Audit one publicly accessible webpage per run for technical SEO, on-page SEO, GEO/AEO readiness, structured data, trust signals, and AI-search citation potential. Generates an evidence-backed Markdown report with verification labels, P0/P1/P2 priorities, and actionable recommendations. For website owners, AI/SaaS teams, e-commerce businesses, creators, SEO consultants, and marketers. Supports English and Simplified Chinese. No API key or paid SEO subscription is required. Live analysis depends on your AI agent’s webpage-access capability. Free version: one public webpage per run. It does not include full-site crawling, batch analysis, authenticated pages, scheduled monitoring, historical comparisons, or proprietary keyword-volume, ranking, traffic, and backlink data. Results do not guarantee indexing, rankings, traffic, revenue, or AI-search citations. Created by ENHE AI Studio https://www.enhe-tech.com.cn/
owasp-security-reviewer
Review code against the OWASP Top 10 and get a prioritized, evidence-based report. Flags injection, broken access control, cryptographic failures, SSRF, and misconfiguration, with the exact file and line, the exploit scenario, and a concrete fix for each finding. Includes severity scoring and a remediation checklist. No account, no API key.
Notice Trap Lite
Notice Trap Lite is a free renewal-risk triage for SaaS stacks. Paste a vendor list (and optional renewal window) and get a short memo with honest LEAD vs opportunity labels. It does not write Google Sheets, pull Ramp, or draft vendor emails. Use when the user has SaaS tools renewing in ≤90 days, wants a one-page renewal risk read before a full vendor-ops pack, or asks an agent to triage renewals / unused seats / renegotiation candidates from a pasted list. One-question intake (spend source? window days?). Next: Notice Trap Pack ($24) — inventory, scoring, approval-ready counters: https://agentpowers.ai/skills/notice-trap-pack
claw-auditor
Audit an OpenClaw or ClawHub skill before you install it. Statically inspects the package for red flags: unsafe shell execution, exfiltration paths, over-broad file access, prompt-injection patterns, obfuscated payloads, and missing provenance. Returns a risk score with per-finding evidence and a clear install or reject verdict.
SKILL.md Validator / Linter
A free, instruction-only SKILL.md validator and linter for agent skills. Checks structure, frontmatter, required sections, and common pitfalls, then returns a clean PASS / REVISE / FAIL verdict with fixes.
CRA Mileage Log Skill for AI Agents
Help your AI assistant keep a CRA-style mileage log for a Canadian vehicle: date, purpose, destination, start and end odometer, and kilometres for each trip, with totals, business-use % and per-km allowance figures from the Tallyfern Rates MCP, each with its source and as-of date. Not tax advice.
Agent Skill Sampler - GitHub PR Review + Gemini Image Generation
Two free agent skills in one: read-only GitHub (PRs awaiting your review, CI status, issues and notifications through your existing gh CLI login, no keys) and nano-banana-pro image generation and editing with Google Gemini 3 Pro Image. Works in Claude Code, Cursor and any agent that loads SKILL.md skills.
X Article: From Notes to a Publish-Ready Package
Plan, write, and package a long-form X Article (the x.com long-form format with title, cover image, headings and inline images) from notes, research, or a rough draft. Covers angle and hook, section outline with image slots, drafting, an evidence pass that removes unsupported numbers, a plain-style edit, alt text, and the promo post. Use for X Articles, not single posts or numbered threads. Made by Dani Acosta, follow on X: https://x.com/DaniAcostaAI
AEO / GEO Site: Build and Audit for AI Answer Engines
Audit or build a website so it is found by search engines and understood, quoted, and cited by AI answer engines (ChatGPT search, Perplexity, Claude, Copilot, Gemini, Google AI Overviews). Use when shipping or reviewing pages, writing schema.org JSON-LD, robots.txt, sitemap.xml or llms.txt, scoring a site for AEO/GEO readiness, or turning an audit into a prioritized fix list. Made by Dani Acosta, follow on X: https://x.com/DaniAcostaAI
era1-offer-writer
Draft a short, buyer-facing freelance offer from a job brief. Use when writing bids, proposals, or fixed-scope service quotes.
Orchestrator Verify Agent Deliverables
Scorecard skill to fact-check Hermes/Agy/Grok outputs before shipping or claiming money moved — blocks confabulation and fake payouts.
BrandLoom: Project Logos & Covers
Need a logo and cover that feel like they belong to the same project? BrandLoom helps you turn your project idea into a clear, consistent visual set. It reads the project information, files, links, and brand materials you provide, then asks one simple question at a time. You choose the wording, style, fonts, characters, layout, language, and image size before anything is created. BrandLoom can make a square project logo image, a wide cover, and a GitHub social preview in Chinese, English, or both. It can reuse your existing company logo and approved brand assets, keep important text clear, save new versions without replacing old ones, and help with later edits, translations, and style changes. It is made for solo founders, indie developers, AI builders, creators, and anyone launching a GitHub project, Skill, app, or digital product. Made for Codex. Full image creation requires Python 3.12 or newer and access to Codex's built-in image tool.
kling-3-0
# Kling 3.0 - Pro Pack on RunComfy [runcomfy.com](https://www.runcomfy.com/?utm_source=skills.sh&utm_medium=skill&utm_campaign=kling-3-0) · [docs](https://docs.runcomfy.com/cli/introduction) · [GitHub](https://github.com/agentspace-so/runcomfy-agent-skills/tree/main/kling-3-0) [Kling 3.0](https://www.runcomfy.com/models/kling/kling-3.0) is Kuaishou Technology's third-generation cinematic video model. This skill covers all six Kling 3.0 rendering endpoints on RunComfy: three quality tiers (Standard, Pro, 4K)
momentic-test
# Momentic model Momentic turns structured natural language into browser automation using forked Playwright, CDP, and custom agents. Interactive steps resolve natural-language targets, assertions can use multimodal models, and AI actions can complete broader goals. # Project files and formats Tests use `*.test.yaml`; reusable modules use `*.module.yaml`. Test IDs live in the test file's `id` field. Check `momentic.config.yaml` and test-level metadata before assuming project defaults apply. `fileType: moment
context-engineering
# Context Engineering ## Overview Feed agents the right information at the right time. Context is the single biggest lever for agent output quality — too little and the agent hallucinates, too much and it loses focus. Context engineering is the practice of deliberately curating what the agent sees, when it sees it, and how it's structured. ## When to Use - Starting a new coding session - Agent output quality is declining (wrong patterns, hallucinated APIs, ignoring conventions) - Switching between different
momentic-mobile-test
# Momentic model Momentic Mobile turns structured natural language into native and webview automation on Android emulators and iOS simulators. Interactive steps resolve targets into device actions; assertions can inspect screenshots, native hierarchies, and webview state. # Project files and formats Mobile tests use `*.test.yaml`; modules use `*.module.yaml`. Test IDs live in the test file's `id` field. Check `momentic.config.yaml` and test-level metadata before assuming project defaults apply. `fileType: m
agent-pulse
# Agent Pulse ## Purpose Use the installed `agent-pulse` CLI as the source of truth for local AI-agent activity. The PyPI package is `agentpulse-cli`, while the command remains `agent-pulse`. Prefer running commands and summarizing their output over reading the Agent Pulse source code. Always enable UTF-8 on Windows before running commands because Agent Pulse output contains emoji and box drawing: ```powershell $env:PYTHONUTF8='1' $env:PYTHONIOENCODING='utf-8' ``` If `agent-pulse` is not on PATH, ask before