Hashproof MCP server
Issue verifiable credentials from AI agents for $0.10 via x402. On-chain on Celo, IPFS-pinned.
1 stars35 downloads/wk
Reviews
Write oneNobody has reviewed Hashproof yet.
If you have run it, two minutes of your experience saves the next person an afternoon.
Hashproof tools (4)
write = sends, deletes, buys or postsRead from the package source without running it. The installed server may list more.
get_template_requirementsGet the required field keys and layout definition of an existing HashProof template. Free, no auth. Always call this before issuing with a template_slug you didn't create in this session.
issue_credentialIssue a verifiable credential — COSTS $0.10 USDC (x402) or 1 API-key credit per call. Get explicit human approval before calling. Requires HASHPROOF_API_KEY or HASHPROOF_WALLET_PRIVATE_KEY in the MCP server env (if neither is set, the error explains how to set up a dedicated wallet). The body follows the HashProof issueCredential schema: issuer{display_name,slug}, platform{display_name,slug}, hold
preview_templateRender a watermarked preview PDF from an INLINE template definition — free, nothing stored, no payment. Use this to iterate field positions (x, y, width, font_size) until the layout is right, then show the final PDF to the human for approval before issuing. page_width/page_height MUST match the background image's pixel dimensions. The PDF is written to a local file; if you can read PDFs, inspect i
verify_credentialVerify an issued credential against three independent sources (database, IPFS, blockchain). Free, no auth. Returns status (active/revoked/expired/not_found), issuer verification and the credential data.
Public scan report
scanner v0.1.9 · 2026-09-25 · same rubric, same numbers if you re-run it
- Code scan2 source files scanned25/25
- –Live reliabilityno gateway calls yet and no remote to proben/a
- –Tool poisoningtools not inspected (local package is not executed); not countedn/a
- Auth qualitystatic API keys via environment variables6/15
- Maintenancelast push 48 days ago12/15
- Maintainer identityregistry namespace matches repository owner; GitHub account older than a year8/10
How Hashproof compares
Against every graded MCP server, from our own scans and uptime checks. Updated after each scan run.
- Safety score
- 78/100
- higher than 57% of 48,255 servers
- Tools
- 4
- the typical server has 7
- Tools that change things
- 0%
- typical: 0%. Fewer write tools means less an agent can break
- GitHub stars
- 1
- more than 65% of 28,688 servers
- Downloads a week
- 35
- more than 25% of 12,191 servers with a package
What its tools let an agent do: 4 read
Install Hashproof in Claude Code, Cursor or VS Code
Runs npx -y hashproof-mcp on your machine. Read the scan report first; the gateway never runs local packages.
claude mcp add hashproof -- npx -y hashproof-mcp
What the publisher says
From the Hashproof repository's README, as published. We do not edit it. Read it on GitHub
HashProof
Verifiable credentials with IPFS backup, on-chain registry, and pay-per-call API via x402.
What it is
HashProof is a credential issuance API. Organizations and Individuals (entities) issue verifiable credentials to people. Each credential is:
- Stored as a JSON in the database and backed up on IPFS (Pinata)
- Registered on-chain in the CredentialRegistry contract on Celo
- Verifiable via a public URL: https://hashproof.dev/verify/:id
Two ways to pay: micropayment in USDC via x402 (Base or Celo) — no API keys, no subscriptions — or a prepaid API key for enterprise clients (no crypto needed). AI agents can call the API directly using a funded wallet.
Project structure
backend/ Express API — credential issuance, verification, payments
frontend/ React app — landing page, credential verification, entity pages
contracts/ CredentialRegistry smart contract (Celo)
mcp/ MCP server (hashproof-mcp on npm) — use HashProof from any AI agent
skills/ Agent skill (npx skills add csacanam/hashproof)
docs/ Architecture and flow documentationDeployed contracts
Paid API endpoints
Payment is in USDC on Base or Celo via x402. The client signs an off-chain authorization — no gas required.
Free endpoints
For AI agents
HashProof is fully agent-usable — no account, no human onboarding. Three ways in:
- MCP server, local or remote (listed on the official MCP registry as io.github.csacanam/hashproof):
# local (stdio) — pays per credential with your own wallet via x402, or an API key
claude mcp add hashproof -- npx -y hashproof-mcp
# remote (Streamable HTTP) — nothing to install, API key required to issue
claude mcp add --transport http hashproof https://api.hashproof.dev/mcpTools: previewtemplate (free — iterate custom layouts), issuecredential ($0.10 via x402 or API key), verifycredential and gettemplate_requirements (free). The remote server has no wallet, so it can't do x402 — it needs Authorization: Bearer . See mcp/README.md.
- Agent skill: npx skills add csacanam/hashproof — or read it at hashproof.dev/skill.md.
- Plain HTTP + x402: POST /issueCredential returns a 402 challenge any x402-v2 client can pay (USDC on Base or Celo). LLM index: hashproof.dev/llms.txt.
Agent identity metadata (ERC-8004 registration-v1): hashproof.dev/metadata.json.
Quick start
See backend/README.md and frontend/README.md for setup instructions. See docs/README.md for the full documentation index.
Nothing above is checked by us. What we check is on the safety report.
Hashproof: common questions
- Is Hashproof MCP server safe?
- Mostly: it is graded B (78/100). Read the Hashproof safety report
- How do I install Hashproof?
- It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
- Does Hashproof need an API key?
- Yes. The registry entry asks for
HASHPROOF_API_KEY,HASHPROOF_WALLET_PRIVATE_KEY. - Is Hashproof maintained?
- The last commit was 87 days ago (2026-08-09). The latest release is v0.1.2.
- How does Hashproof compare with other MCP servers?
- safety score 78/100 (higher than 57% of 48,255 servers); tools 4 (the typical server has 7); tools that change things 0% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.