AI Proof of Us
Signed local receipts for AI tasks, with optional claims on Base
- Pricing
- Open source
- Platforms
- CLI
- Runs
- Locally
- Version
- 0.6.1
- Listing updated
- 3 days ago
About AI Proof of Us
AI Proof of Us is an open-source MCP server for ai models and llm tooling. It runs locally on your machine and is started by your AI client.
AI Proof of Us treats a finished AI task as something to sign. The server records a private, signed receipt that stores local hashes instead of raw prompts or outputs, so receipts can be kept without ever claiming a token. Verification of a receipt alone reports host execution as UNKNOWN.
You can try it with npx -y aipou-mcp-server --demo, which creates, verifies and discards one receipt without a wallet or funds. A persistent setup uses --init to make a protected dedicated wallet key and --doctor to run read-only checks. If you want to claim, validator-approved pending receipts may enter a Merkle batch and a contract on Base verifies proofs and mints AIPOU. The project warns to use a dedicated wallet, never a primary one.
The project is explicit about limits: it is experimental, open source and unaudited, makes no promise on price or rewards, does not detect hidden AI use and does not prove task quality. One external developer trial is reported, and it says recurring use is not established.
Install
Run in a terminal
claude mcp add ai-proof-of-us --env AIPOU_AGENT_PRIVATE_KEY='<AIPOU_AGENT_PRIVATE_KEY>' --env AIPOU_CONTRACT_ADDRESS='<AIPOU_CONTRACT_ADDRESS>' --env AIPOU_CLAIMS_ADDRESS='<AIPOU_CLAIMS_ADDRESS>' -- npx -y aipou-mcp-serverAdd to claude_desktop_config.json
{
"mcpServers": {
"ai-proof-of-us": {
"command": "npx",
"args": [
"-y",
"aipou-mcp-server"
],
"env": {
"AIPOU_AGENT_PRIVATE_KEY": "<AIPOU_AGENT_PRIVATE_KEY>",
"AIPOU_CONTRACT_ADDRESS": "<AIPOU_CONTRACT_ADDRESS>",
"AIPOU_CLAIMS_ADDRESS": "<AIPOU_CLAIMS_ADDRESS>"
}
}
}
}Add to ~/.cursor/mcp.json
{
"mcpServers": {
"ai-proof-of-us": {
"command": "npx",
"args": [
"-y",
"aipou-mcp-server"
],
"env": {
"AIPOU_AGENT_PRIVATE_KEY": "<AIPOU_AGENT_PRIVATE_KEY>",
"AIPOU_CONTRACT_ADDRESS": "<AIPOU_CONTRACT_ADDRESS>",
"AIPOU_CLAIMS_ADDRESS": "<AIPOU_CLAIMS_ADDRESS>"
}
}
}
}Add to .vscode/mcp.json
{
"servers": {
"ai-proof-of-us": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"aipou-mcp-server"
],
"env": {
"AIPOU_AGENT_PRIVATE_KEY": "<AIPOU_AGENT_PRIVATE_KEY>",
"AIPOU_CONTRACT_ADDRESS": "<AIPOU_CONTRACT_ADDRESS>",
"AIPOU_CLAIMS_ADDRESS": "<AIPOU_CLAIMS_ADDRESS>"
}
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"ai-proof-of-us": {
"command": "npx",
"args": [
"-y",
"aipou-mcp-server"
],
"env": {
"AIPOU_AGENT_PRIVATE_KEY": "<AIPOU_AGENT_PRIVATE_KEY>",
"AIPOU_CONTRACT_ADDRESS": "<AIPOU_CONTRACT_ADDRESS>",
"AIPOU_CLAIMS_ADDRESS": "<AIPOU_CLAIMS_ADDRESS>"
}
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"ai-proof-of-us": {
"command": "npx",
"args": [
"-y",
"aipou-mcp-server"
],
"env": {
"AIPOU_AGENT_PRIVATE_KEY": "<AIPOU_AGENT_PRIVATE_KEY>",
"AIPOU_CONTRACT_ADDRESS": "<AIPOU_CONTRACT_ADDRESS>",
"AIPOU_CLAIMS_ADDRESS": "<AIPOU_CLAIMS_ADDRESS>"
}
}
}
}Configuration
AIPOU_AGENT_PRIVATE_KEYOptional inline private key for a dedicated farming wallet. Prefer AIPOU_AGENT_KEY_FILE and never use a primary wallet.AIPOU_AGENT_KEY_FILEPath to the protected dedicated-wallet key created by aipou-mcp --init.AIPOU_CONTRACT_ADDRESSAIPOU token contract address on the configured network.RequiredAIPOU_CLAIMS_ADDRESSAIPOU claims contract address on the configured network.RequiredAIPOU_RPC_URLOptional Base-compatible JSON-RPC endpoint.AIPOU_DATA_DIROptional local directory for receipt and collector state.
Tools this server provides (2)
Taken from the project documentation; the running server may differ.
- complete_ai_task
- Record a signed receipt for a completed AI task.
- export_ai_receipts
- Export persisted receipts, including recovery by nonce.
What it does
- Signed receipts
- Record private signed receipts for authorised AI tasks.
- Hash-only storage
- Keeps local hashes rather than raw prompts or outputs.
- Disposable demo
- Run --demo to create and discard one receipt.
- Setup diagnostics
- --init and --doctor create and check a dedicated wallet.
- Optional claims
- Validator-reviewed receipts can be claimed on Base.
- Claim recovery
- Recover a persisted receipt by nonce after a lost response.
Good for
- Attach a verifiable receipt to an agent task
- Check pending and claimed AIPOU before choosing to claim
- Evaluate a receipt scheme for an agent framework
- Keep local task receipts without claiming tokens
Pricing
Open source
Open source; claims involve a token on Base
Prices change. Check 0xddneto.github.io before you buy.
AI Proof of Us FAQ
Does it store my prompts?
The protocol stores local hashes instead of raw prompts or outputs.
Do I need a wallet to try it?
No. The --demo command needs no wallet, funds or setup.
Does it prove the AI work was good?
No. It does not prove task quality or detect hidden AI use.
Is it audited?
The project describes AIPOU as experimental, open source and unaudited.
Is AI Proof of Us free?
Yes. AI Proof of Us is open source, so you can use and self-host it for free.
How do I add AI Proof of Us to Claude Code?
Run this command in your terminal: claude mcp add ai-proof-of-us --env AIPOU_AGENT_PRIVATE_KEY='<AIPOU_AGENT_PRIVATE_KEY>' --env AIPOU_CONTRACT_ADDRESS='<AIPOU_CONTRACT_ADDRESS>' --env AIPOU_CLAIMS_ADDRESS='<AIPOU_CLAIMS_ADDRESS>' -- npx -y aipou-mcp-server
What are the best alternatives to AI Proof of Us?
The closest alternatives to AI Proof of Us are Claude Code Ultimate Guide, signafAIde, Toolkien, Lugh and HypePaper.
Alternatives to AI Proof of Us

Claude Code Ultimate Guide
MCP serverMCP server for searching the Claude Code Ultimate Guide from any MCP client
Runs locally- Open source
- 6.5K GitHub stars
- AI models and LLM tooling

signafAIde
MCP serverLocal MCP server to set up, verify and check coverage of signed agent action records
Runs locally- 5 tools
- Freemium
- Security

Toolkien
MCP serverOne MCP gateway for all your MCP servers, with code mode, permissions and audit
Not checked yet- 3 tools
- Remote
- Open source
- AI models and LLM tooling
Lugh
MCP serverWrite, check and dry-run Lugh agent programs with policy, budgets and approvals
Runs locally- 5 tools
- Open source
- AI models and LLM tooling
HypePaper
MCP serverResearch papers, benchmarks, citations and topic experts for AI assistants
- 58 tools
- Remote
- Freemium
- AI models and LLM tooling

One compact entry point that finds, describes and runs hundreds of pay-per-call data tools
- 3 tools
- Remote
- Freemium
- Developer tools