About evermint-mcp
evermint-mcp records AI agent actions in tamper-evident, cryptographically timestamped receipts. It is for teams needing to inspect or verify individual records and hash-chained action histories.
Key features
- 01Mint timestamped action records
- 02Verify individual records
- 03List records with filters
- 04Retrieve full record payloads
- 05Verify hash-chained record sequences
Use cases
- 01Create an audit record for an agent decision
- 02Check the integrity of a record by ID
- 03Review an organization's action records
- 04Validate links across a sequence of records
MCP support
Exposes tools for creating, retrieving, listing, and verifying action records to MCP clients. Minting and organization record access require an API key; individual and chain verification are public.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
npx -y evermint-mcpEVERMINT_API_KEYSecretYour EverMint API key (starts with EVR-sk_). Get one at https://evermint.app/api-keysLocal servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.
Tools
5Saved tool list · updated 2w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| payload | object | No | The state, decision, or context to preserve. Any JSON object. |
| agent_id | string | No | Identifier for this agent or system |
| timestamp | string | No | ISO 8601 datetime or 'auto'. Defaults to auto. |
| action_type | string | Yes | Short label for the action being recorded (e.g. transaction_approved, decision_made, data_accessed) |
Details
From the project READMEevermint-mcp
Tamper-evident receipts for AI agent actions. The notary layer for agent-to-agent transactions.
Install
npm install -g evermint-mcp
Claude Desktop config
Add to your claude_desktop_config.json:
{
"mcpServers": {
"evermint": {
"command": "evermint-mcp",
"env": {
"EVERMINT_API_KEY": "EVR-sk_your_key_here"
}
}
}
}
Get an API key at https://evermint.app/api-keys
- Add to your Claude desktop config (
~/.claude/claude_desktop_config.json):
{
"mcpServers": {
"evermint": {
"command": "npx",
"args": ["-y", "evermint-mcp"],
"env": {
"EVERMINT_API_KEY": "EVR-sk_your_key_here"
}
}
}
}
- Restart Claude. The tools appear automatically.
Tools
evermint_mint (API key required)
Mint a tamper-evident, cryptographically timestamped record of an action.
evermint_mint({
agent_id: "billing-agent-1",
action_type: "decision_made",
payload: { decision: "approved", confidence: 0.97 }
})
evermint_verify (public)
Verify a single record by Record ID.
evermint_verify({ record_id: "EVR-12C5N1A2" })
evermint_list_records (API key required)
List records for the authenticated org with optional filters. Returns metadata only (no payloads).
evermint_list_records({
agent_id: "billing-agent-1",
action_type: "decision_made",
since: "2026-01-01T00:00:00Z",
until: "2026-05-01T00:00:00Z",
limit: 50
})
All filters are optional. limit defaults to 50 (max 200).
evermint_get_record (API key required)
Fetch the full contents of a single record, including its original payload. Returns 403 if the record exists but belongs to a different organization.
evermint_get_record({ record_id: "EVR-12C5N1A2" })
evermint_verify_chain (public)
Verify a sequence of hash-chained records. Confirms each hash matches its content, and each chain_link correctly references the prior record.
// Explicit list, oldest first
evermint_verify_chain({
record_ids: ["EVR-AAAAAAAA", "EVR-BBBBBBBB", "EVR-CCCCCCCC"]
})
// Or walk forward from a starting point
evermint_verify_chain({
start_record_id: "EVR-AAAAAAAA",
length: 10
})
Returns per-record hash_valid, chain_link_valid, plus an overall chain_intact: true|false.
Verify any record
evermint.app/verify?id=EVR-XXXXXXXX
Publishing
To publish to npm as evermint-mcp:
package.jsonname:"evermint-mcp"binentry:{ "evermint-mcp": "./dist/server.js" }- Add to Anthropic MCP registry at modelcontextprotocol.io/registry
Contact
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.