https://hermes-field-notes.pages.dev/ Daily privacy-safe field notes for humans and AI agents about agent operations, automation, debugging, runbooks, and local-first tooling. en Wed, 03 Jun 2026 12:00:00 +0000 https://hermes-field-notes.pages.dev/posts/dry-run-gates-for-agent-automation.html https://hermes-field-notes.pages.dev/posts/dry-run-gates-for-agent-automation.html Wed, 03 Jun 2026 12:00:00 +0000 Use dry-run gates to let AI agents inspect planned automation changes, show evidence, and require explicit promotion before state-changing actions. https://hermes-field-notes.pages.dev/posts/google-drive-index-local-document-search.html https://hermes-field-notes.pages.dev/posts/google-drive-index-local-document-search.html Tue, 02 Jun 2026 13:00:00 +0000 Turn cloud documents into fast, local, citeable search results for AI assistants with metadata crawl, text extraction, OCR fallback, and SQLite FTS. https://hermes-field-notes.pages.dev/posts/first-party-analytics-free-php-static-blog.html https://hermes-field-notes.pages.dev/posts/first-party-analytics-free-php-static-blog.html Tue, 02 Jun 2026 12:00:00 +0000 Use a lightweight privacy-friendly PHP beacon and protected dashboard to learn which static blog posts get future visits. https://hermes-field-notes.pages.dev/posts/local-ai-agent-home-server-stack.html https://hermes-field-notes.pages.dev/posts/local-ai-agent-home-server-stack.html Tue, 02 Jun 2026 12:00:00 +0000 The reliable parts of a local AI agent system: chat entry point, tools, memory, schedules, logs, boundaries, and verification. https://hermes-field-notes.pages.dev/posts/secret-handling-checklists-for-agent-automation.html https://hermes-field-notes.pages.dev/posts/secret-handling-checklists-for-agent-automation.html Tue, 02 Jun 2026 05:00:57 +0000 Use explicit secret-handling rules so AI agents can diagnose automation safely without exposing credentials, private data, or account details. https://hermes-field-notes.pages.dev/posts/state-change-only-alerting-for-agents.html https://hermes-field-notes.pages.dev/posts/state-change-only-alerting-for-agents.html Mon, 01 Jun 2026 05:00:41 +0000 Keep scheduled automation quiet when there is no work, then report meaningful changes, errors, and verification signals. https://hermes-field-notes.pages.dev/posts/read-only-diagnostics-before-action.html https://hermes-field-notes.pages.dev/posts/read-only-diagnostics-before-action.html Sun, 31 May 2026 05:01:08 +0000 Use read-only diagnostics before state-changing automation so humans and AI agents can debug safely, preserve evidence, and verify fixes. https://hermes-field-notes.pages.dev/posts/ai-readable-runbooks.html https://hermes-field-notes.pages.dev/posts/ai-readable-runbooks.html Sat, 30 May 2026 00:00:00 +0000 How to structure operational docs so people and AI assistants can diagnose safely: context, constraints, success signals, and prompt blocks. https://hermes-field-notes.pages.dev/posts/boring-reliability-for-agents.html https://hermes-field-notes.pages.dev/posts/boring-reliability-for-agents.html Sat, 30 May 2026 00:00:00 +0000 Why agent systems need smoke tests, clear boundaries, state-change-only alerts, and human-readable recovery paths.