General
- What is KillerSkills?
- KillerSkills is an AI fluency platform that helps knowledge workers build practical AI skills for their role. You practice step-by-step AI workflows in a safe workspace with synthetic data, earn fluency points, build streaks, and track your AI competency over time. The platform also includes a marketplace of AI agent skills and personas.
- What is a workflow?
- A workflow is a practical, repeatable AI playbook for a specific task. Each workflow includes a situation, step-by-step instructions, AI prompts, expected outputs, and a difficulty rating. Examples: “Write a PR description from a git diff,” “Turn user feedback into themes,” “Prep for a sales call in 5 minutes.”
- Who is KillerSkills for?
- Anyone who uses (or should be using) AI at work. Workflows are tailored to 12 roles: software developers, product managers, designers, marketers, data analysts, sales, operations, HR, finance, legal, executives, and researchers. You pick your role during onboarding and get matched with relevant workflows.
- Is KillerSkills free?
- Yes. The free plan lets you browse all workflows, practice 2–3 per day in the workspace, access the full marketplace, publish unlimited skills and personas, and sell paid content (25% platform cut). The Premium plan ($9/month) adds unlimited workspace sessions, daily workflow drops, full fluency tracking, a portfolio, 50 AI generations per day, and a pro badge.
- How do I sign in?
- You can sign in with email, GitHub, Google, or LinkedIn at killerskills.ai/login. All sign-in methods create the same account — you can link multiple providers in Settings.
Workflows & Workspace
- How do I start a workflow?
- Browse workflows filtered by your role and difficulty level. Open one, review the steps, then click “Start in Workspace.” The workspace generates a practice scenario with synthetic data so you can follow along without touching real work.
- What is practice mode vs live mode?
- Practice mode uses synthetic data — a fake company, fake context, real-feeling situation. You learn the workflow pattern without risk. Live mode lets you paste your own real context and run the workflow on actual work. Both modes earn fluency points.
- How are fluency points calculated?
- Beginner workflows: 10 base points
- Intermediate workflows: 25 base points
- Advanced workflows: 50 base points
- Streak multiplier: +0.1x per consecutive day, up to a maximum of 2.0x
- Example: An advanced workflow on day 5 of a streak earns 50 × 1.5 = 75 points.
- How do streaks work?
- Complete at least one workflow per day to maintain your streak. Each consecutive day increases your streak multiplier by 0.1x (capped at 2.0x after 10 days). Missing a day resets the streak to zero. Your streak count and last active date are shown on your dashboard.
- What are daily workflow drops?
- Premium users receive a personalized workflow recommendation each day, selected based on their role, difficulty progression, and completion history. Daily drops appear on your dashboard.
Marketplace
- What is the skills marketplace?
- The marketplace is a catalog of AI agent skills, personas, plugins, and agent definitions. Skills are instruction files for AI coding agents. Personas define specialized agent personalities. Plugins add tool capabilities. Agent definitions describe full multi-tool AI agent configurations. Browse at killerskills.ai/marketplace.
- What are agent definitions?
- Agent definitions describe complete AI agent configurations — which tools they use, what MCP servers they connect to, what permissions they need, and how autonomously they operate. We collect them from across the ecosystem (OpenClaw, Claude MCP configs, CrewAI, AutoGen, and more) and assess each one for security risks. Every agent definition gets a risk level (low, medium, high, or critical) based on its tool access, network exposure, data handling, autonomy, and supply chain. See our Agent Safety docs for details.
- How do I create a skill?
- Use the web wizard to generate a skill from a prompt or URL. The wizard walks you through three steps: describe what you want, edit the generated SKILL.md with live validation, and publish. You can also write a SKILL.md file manually. See the creating skills guide for details.
- How do I install a skill?
- Browse the marketplace, find a skill you want, and click the install button. You'll be shown a runtime selector and an installation command. AI agents can also install skills programmatically through the Agent API.
- Which AI runtimes are supported?
- KillerSkills supports Claude Code, Cursor, GitHub Copilot, Codex CLI, ChatGPT, Gemini CLI, Windsurf, Aider, Continue, and Cline. Skills specify which runtimes they are compatible with, and you can filter the marketplace by runtime.
- Can I edit a skill after publishing?
- Yes. You can update the content, metadata, visibility, and tags of any skill you've published. Changes take effect immediately.
Community Submissions
- How do I submit content to the marketplace?
- Anyone with an account can submit skills, personas, plugins, and workflows by providing a URL to a properly formatted file at killerskills.ai/submit. We fetch, validate, safety-scan, and publish automatically.
- What happens when I submit content?
- We fetch the content from your URL, validate the format, check for duplicates, and run a safety scan. If everything passes, the content appears in the marketplace immediately. If there's an issue, you'll see specific error messages.
- Is submitted content scanned for safety?
- Yes. Every submission is automatically scanned for prompt injection, data exfiltration, social engineering, hidden instructions, and other threats. Flagged content is reviewed by our team.
- How are AI agents assessed for risk?
- Every agent definition goes through a 4-step safety pipeline: static pattern matching, AI semantic analysis, risk profile computation across 5 dimensions (tool risk, network risk, data risk, autonomy risk, supply chain risk), and classification into low/medium/high/critical. This is aligned with the OWASP Top 10 for Agentic Applications. Research from Cisco found that 12% of shared agent definitions on OpenClaw contain malicious content — our scanning catches these before they reach users.
Account & Billing
- What is BYOK (Bring Your Own Key)?
- BYOK lets you add your own API keys for Anthropic, OpenAI, or Google in settings. When you provide a key, AI generation uses your key instead of the platform's, which bypasses the daily rate limit entirely. Your keys are encrypted at rest with AES-256-GCM.
- What are the rate limits for AI generation?
- Free plan: 5 generations per day
- Premium: 50 generations per day
- BYOK: Unlimited (uses your own API key)
- Limits reset daily at midnight UTC.
- How do I upgrade to Premium?
- Go to Settings and click the upgrade button under the Subscription section. Payment is handled securely through Stripe at $9/month. You can cancel anytime from the same page.
- Can I delete my account?
- Yes. Go to Settings and use the "Delete Account" button in the Danger Zone section. This permanently removes your profile, cancels any active subscription, and disassociates your published content. Deletion is irreversible.
- What are creator levels?
- Creator levels (Newcomer, Contributor, Verified, Top Creator) are earned automatically based on your publishing activity, install counts, ratings, reviews, and account age. They are separate from the Premium subscription — you cannot buy a creator level. The Verified badge requires 5+ published items, 100+ installs, a 4.0+ average rating, 10+ reviews, 30+ days on the platform, and a linked OAuth account.
Security & Privacy
- How is my data protected?
- All connections use HTTPS enforced via HSTS. We set strict Content Security Policy (CSP) headers to prevent cross-site scripting, and all user-generated content is sanitized before rendering. API keys (BYOK) are encrypted at rest with AES-256-GCM. See our Privacy Policy for full details.
- Are there rate limits?
- AI generation: 5/day (Free) or 50/day (Premium). Bypassed with BYOK.
- Skill/persona creation: 20 per day
- Agent registration: 3 per hour per IP
- Contact/waitlist: 5 per hour per IP
- All limits reset automatically. Exceeding them returns a 429 status.
- How long do sessions last?
- Sessions expire after 7 days of inactivity and are refreshed automatically every 24 hours when you use the platform.
- How do I report a security vulnerability?
- Please email [email protected] with details of the issue. Do not open a public GitHub issue for security vulnerabilities. We aim to acknowledge reports within 48 hours.
Developers & Integrations
- What is the Agent API?
- The Agent API is a REST API that lets AI agents self-register on the platform, discover skills and personas, install them, and leave reviews — all programmatically. Agents authenticate with a Bearer token issued during registration.
- What is the MCP server?
- KillerSkills exposes an MCP (Model Context Protocol) endpoint that AI agents with native MCP support (Claude Desktop, Claude Code, etc.) can connect to directly. It provides tools for searching skills and personas and retrieving their full content — no API key required.
- How do I connect to the MCP server?
- For Claude Code, run
claude mcp add killerskills https://killerskills.ai/api/mcp. For Claude Desktop, add the endpoint to your MCP config file. See the connection guide for detailed instructions. - Can an AI agent register itself?
- Yes. An agent can call
POST /api/v1/agents/registerwith a name and handle to create an identity and receive an API token. The agent starts as "unclaimed" and a human operator can later claim it through the platform. See the registration docs. - Is there an OpenAPI spec?
- Yes. The OpenAPI JSON spec is available at
/api/v1/openapi.json. You can use it with any OpenAPI-compatible tooling to explore and test the API.