AgentSeal

@JoeyBrar

About AgentSeal

Every agent action is recorded in a SHA-256 hash chain. With this, you can actually prove to clients that your agent did what it said it did. Record, query, verify, and export agent activity.

#security #audit #compliance #logging #ai-agents #agents

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "agentseal": {
      "command": "npx",
      "args": [
        "-y",
        "agentseal-mcp"
      ],
      "env": {
        "AGENTSEAL_API_KEY": "as_sk_your_key_here"
      }
    }
  }
}

Tools

Record an agent action to the AgentSeal audit hash trail. Call this after every significant action (sending emails, modifying files, running queries, making API calls) to create a cryptographically sealed record of what happened and why.

Look up previously recorded actions from the audit trail. Use this to check what actions have been taken, verify history, or recall past decisions.

Verify the integrity of the audit trail hash chain. Each entry's SHA-256 hash includes the previous entry's hash — if any record was modified, the chain breaks and this will report where.

Overview

What is AgentSeal?

AgentSeal is an MCP server that provides an immutable audit trail for AI agents, using SHA-256 hash chaining to ensure data integrity. It is designed for multi-agent tracking, on-demand verification, and export, built to meet EU AI Act compliance requirements.

How to use AgentSeal?

Set up in one command: npx agentseal-mcp with your API key. Use the server’s three tools — record_action, query_actions, and verify_chain — to record, look up, and verify agent actions. A Python SDK (pip install agentseal-sdk) is also available for direct integration.

Key features of AgentSeal

One-command setup (npx agentseal-mcp) with API key
On-demand chain integrity verification
Multi-agent tracking under one organization
Export full trail as JSON or CSV
Python SDK for direct integration (agentseal-sdk)
SHA-256 hash chain — tamper-evident records
EU AI Act ready for high-risk AI compliance

Use cases of AgentSeal

Record and verify actions of a single AI agent
Monitor multiple agents across an organization
Provide immutable logs for regulatory audits
Export agent activity trails for analysis or reporting
Ensure data integrity in high-risk AI deployments

FAQ from AgentSeal

What makes AgentSeal different from a regular database?

AgentSeal uses a SHA-256 hash chain where each entry’s hash includes the previous one, so any modification breaks the chain. This provides tamper-evident, verifiable integrity that ordinary databases do not guarantee.

What runtime or dependencies does AgentSeal require?

No additional runtime—just run the one-command npx agentseal-mcp with your API key. A Python SDK is available for direct integration if preferred.

Where does the audit data live?

Data is stored and managed via the AgentSeal service; you sign up at agentseal.io for an API key. The server records and retrieves action data through that service.

What transports or authentication does AgentSeal use?

Authentication is handled via your API key passed with the npx command. The server exposes standard MCP tools over stdio.

Are there any known limitations?

The README does not mention specific limits. It supports multi-agent tracking and export in JSON/CSV.

Frequently asked questions

What makes AgentSeal different from a regular database?

What runtime or dependencies does AgentSeal require?

No additional runtime—just run the one-command `npx agentseal-mcp` with your API key. A Python SDK is available for direct integration if preferred.

Where does the audit data live?

Data is stored and managed via the AgentSeal service; you sign up at agentseal.io for an API key. The server records and retrieves action data through that service.

What transports or authentication does AgentSeal use?

Authentication is handled via your API key passed with the `npx` command. The server exposes standard MCP tools over stdio.

Are there any known limitations?

The README does not mention specific limits. It supports multi-agent tracking and export in JSON/CSV.

Comments

Basic information

More Other MCP servers

Maestro

mobile-dev-inc

Painless E2E Automation for Mobile and Web

14.8K

Figs, Relationship Expert

Figs O'Sullivan, LMFT

Couples therapist Figs O'Sullivan (LMFT, 17 years, endorsed by Sue Johnson, creator of EFT) inside your AI. Ask relationship questions answered from his clinical writing, name the pattern in a fight, get a three-step rep