https://jongerius.solutions/ Recent content on Otto Jongerius Hugo en-us &copy; 2026 Otto Jongerius | <a href="https://gohugo.io" target="_blank">Hugo</a> Mon, 13 Apr 2026 09:00:00 +1200 https://jongerius.solutions/post/auditing-github-mcp-agent-receipts/ Mon, 13 Apr 2026 09:00:00 +1200 https://jongerius.solutions/post/auditing-github-mcp-agent-receipts/ <p>Somewhere between Kyoto temples I shipped a signing proxy for <a href="https://agentreceipts.ai">Agent Receipts</a> — an open protocol that gives every AI agent action a cryptographically signed audit trail. The idea is simple: when an agent acts on your behalf, you should be able to prove what happened. Not just logs. Proof.</p> <p><img src="https://jongerius.solutions/images/post/shinkansen-agent-receipts.jpg" alt="Working on Agent Receipts from the Shinkansen"></p> <p>This week I used it to audit itself.</p> <hr> <h2 id="the-problem-with-mcp-tool-calls">The problem with MCP tool calls</h2> <p>When you give Claude access to GitHub via the MCP server, it can create issues, push files, open pull requests. It acts on your behalf. And when it&rsquo;s done, it tells you what it did.</p> https://jongerius.solutions/post/your-ai-agent-just-sent-an-email/ Fri, 03 Apr 2026 09:00:00 +1200 https://jongerius.solutions/post/your-ai-agent-just-sent-an-email/ <p>Last week, I asked an AI agent to clean up some files in a project directory. It did a great job — renamed a few things, deleted some stale configs, updated a README. I know this because I watched it happen in my terminal.</p> <p>But if you asked me to <em>prove</em> what it did? To show you an authoritative record of every action, in order, with cryptographic proof that nothing was altered after the fact? I couldn&rsquo;t.</p>