AskpertYour agent is a generalist. Hire an AI expert.
Built by someone who does the work for a living, so you get their judgment, not a generic answer. Drops into Claude Code, Codex, or any harness you already run.
npx @askpert/cli add jane/tax-classifier
Meta Ads AuditorFree · per call
Immigration ChecklistFree · per callAn expert is a method, not a prompt
An expert is one person's method, packaged: the skills they work by, the workflows that sequence them, and the documents behind both. You call it and get the answer back. The files stay on our servers.
Monetize expertise without sharing it
Your skills, workflows and documents run on our servers. A buyer's agent sends a question and gets an answer back. It never receives the files, so you can rent out the thing you would never hand over. You earn per call, not once at download.
Expert = skills + workflows
One expert can carry several skills, plus the workflow that decides which one runs when, over the documents those skills draw on. That sequencing is what a single skill file cannot give you.
Every expert clears a security scan
A skill you copy off the internet runs with your permissions, and nobody reviewed it. Every expert here passes 19 automated checks for security and quality before it can be listed, and the results sit on its listing before you hire.
One command, 72 harnesses
One line in the terminal adds an expert to Claude Code, Codex, or any of the 72 agent harnesses the CLI knows about. Every expert on the marketplace is free to call today, so trying one costs nothing.
From listing to answer in one sitting
No new app to learn. An expert you hire becomes a tool inside the agent you already use.
Hire
Pick an expert on the marketplace. The listing shows its price, its rating, and the 19 checks it passed. Hiring is free and takes one click.
See what a listing tells youAdd
One command wires it into Claude Code, Cursor, or any of 72 harnesses. Prefer config files? Paste four lines of MCP JSON.
The install commandAsk
Call it from your agent like any other tool. It answers with the seller's method, on a model Askpert pays for. No provider key needed.
Your first consultBrowse without an account
Every expert on the marketplace is free to call today. Compare them and read their inspection results before you sign up for anything.
Example listings. Browse the real marketplace to see what is published today.
Meta Ads Auditor
Code Review Expert
SEO Brief Generator
Immigration Checklist
One command. It's in your harness.
The CLI detects your harness and writes the config. Claude Code, Codex, Cursor, and 69 more.
npx works with npm, pnpm, bun, and yarn. No global install, no account setup to try it.
Every expert passes 19 checks. You can read them.
Publishing runs an automated inspection: 10 security checks, 5 readiness checks, and a quality review scored against an 80-point bar. A hard failure blocks the listing. The expert's page shows you exactly what passed.
Read all 19 checks
- No hidden instructions. Scanned for invisible characters, hidden directives, and obfuscated content.
- No data exfiltration. Checked that nothing instructs the expert to send your data to outside URLs.
- No credential harvesting. Verified the expert never tries to read environment secrets or credential files.
- Safe skill code. Statically analyzed any executable skill code for dangerous calls.
- No prompt self-leak. Confirmed the expert won't reveal its own hidden system prompt or rules.
- No harmful directives. Screened for instructions that could produce harmful or illegal output.
- Pinned connections. Every external connection is locked to specific named hosts, with no wildcards.
- No raw network access. Network-enabled server-side code can't open raw sockets, spawn processes, or bypass its declared-host allowlist.
- Valid tool manifests. Every code tool declares a valid, domain-pinned manifest.
- Connection hygiene. Reviewed connection settings for safe defaults.
- Model configured. Backed by at least one working language model.
- Documents ready. Attached knowledge documents are ready for the expert to search at run time.
- Substantive instructions. Instructions are detailed enough to do the job well.
- Complete description. The listing has a meaningful description.
- Price set. A clear per-call price is configured.
- Specific instructions. The expert's instructions are specific, coherent, and purposeful.
- Accurate listing. The marketplace description matches what the expert actually does.
- Clear scope. Inputs, outputs, and boundaries are clearly defined.
- Real expertise. Shows genuine domain depth, not generic prompting.
Have a method worth renting out?
Seal your process into an expert: the documents and checklists you already work from. Buyers get answers through their own agents and never see what's inside. Build it in Creator Studio, or from the harness you already use, with no provider key of your own.
Fair questions
How is this different from a skill marketplace?
Most skill marketplaces sell you the skill itself. You download the SKILL.md folder onto your machine, so from then on it can be copied, forwarded, or resold, and the author is paid once and loses control of their know-how. Askpert works the other way around: the skill stays sealed on our servers. You never download it. Your own agent calls it over MCP as a tool, and the instructions, knowledge, and code never cross the wire. The seller keeps their method and earns per call, and every listing clears a 19-check inspection before it can go live.
Do I need my own API keys?
No, not to start. Every expert includes a free default model, DeepSeek V4 Flash. Askpert pays for it, so you can call any expert with just your Askpert API key. Bring your own LLM API key only when you want to run one of the models a seller suggests instead: you pass that model on your first consult, and the run bills to your provider account at your provider's normal rates. Add keys once under Settings, API keys, on the Buyer tab, and every expert reuses them.
Which model providers can I use?
Every expert runs on the included default model with no key from you: DeepSeek V4 Flash, or GPT-5.6 Luna if the expert accepts file attachments, since V4 Flash is text-only. Sellers can also suggest OpenAI, Gemini, or OpenRouter models; to run one of those, add that provider's LLM API key under Settings, API keys. Some experts also search the web through Exa or Parallel, but those run on the seller's key, so there is nothing extra for you to add.
What does hiring cost?
Two things, and they stay separate. First, the expert's own per-call price. The seller sets it in credits, and every expert listed today is free at 0 credits. If a seller does price an expert, that price shows on the listing before you commit. Second, model tokens. On the included default model Askpert covers them. If you choose a model the seller suggested, that run uses your own LLM API key at your provider's normal rates, so the token spend lands on your account and is never bundled into anyone else's bill. Browsing itself needs no account.
Can an expert read my files or leak my data?
An expert only ever receives what you choose to send it. Each listing declares up front whether it accepts files at all, and when it does, your own harness gathers them locally and passes along only what is needed, so your raw files do not leave your machine on their own. If an expert runs code over data, that code runs in an isolated server sandbox with no internet access unless the listing names specific domains, and it can never reach back into your local machine. Every reply also carries a safety notice and machine-readable risk flags, so your harness can pause anything sensitive.
Who can see what I send to an expert?
It depends which model runs. On the included default model the call goes out on Askpert's own OpenRouter account, so the conversation is handled under OpenRouter's terms and those of the upstream provider serving DeepSeek V4 Flash. If you run a model the seller suggested, the call goes out on your own LLM API key and the conversation lives in your provider account under that provider's terms. Either way the seller never sees your messages, nothing you send reaches other buyers, and usage tracking stores counts and dates, not content.
Whose keys pay for what?
Askpert pays for the included default model every expert runs on, plus the publish inspection and the generated avatar art. Buyers bring an LLM API key only to run a model a seller suggested instead of the default. Sellers bring keys for the models they choose to build with; with no key on file, drafting and test runs fall back to the included default model, which Askpert funds. Nobody's optional model usage is ever bundled into someone else's bill.
What happens to an expert I hired if the seller changes or removes it?
You run a frozen copy of what passed inspection, byte for byte. A seller's later edits change nothing for you until they re-inspect and re-publish. If a seller unpublishes, new calls stop; nothing you already received is touched.
I'm a seller. Can a buyer extract my prompts or documents?
Buyers get answers, not contents. Instructions, skills, documents, and code stay server-side; the API returns name, description, and reachable domains for a code tool, nothing more. An anti-extraction guard wraps every expert, and the inspection includes checks that it won't reveal its own instructions.
The marketplace is open.
Hire an expert into your agent this afternoon, or seal your own and list it. One account does both.