Facio Blog

Practical notes on human-reviewed AI agents.

Payload-powered product notes, security writing, HITL patterns, and operational guidance from the Facio runtime: long sessions, Placet approvals, audit trails, memory, providers, channels, tools, and Docker-first operations.

Showing 56-60 of 78 articles in Product.

Product

Facio's Tool Architecture: How a Composable Runtime Turns LLM Calls Into Real-World Actions

Jun 12, 2026Product

Facio's Tool Architecture: How a Composable Runtime Turns LLM Calls Into Real-World Actions

An LLM that can only generate text is a research tool, not a worker. The real power of AI agents comes from tools — the structured interface that turns model reasoning into real-world actions. Facio's tool architecture is built for composability: every tool follows the same patterns, every tool respects HITL boundaries, and every tool can be combined with every other tool to build workflows that no single tool enables. Here's how the architecture works.

Product

Facio's Message & Upload Tools: How AI Agents Deliver Files and Conversations Across Any Channel

Jun 11, 2026Product

Facio's Message & Upload Tools: How AI Agents Deliver Files and Conversations Across Any Channel

An AI agent that produces a perfect report but can't deliver it to a human is a journal, not a worker. Facio's message and upload_file tools close the loop — sending text, images, videos, and documents to any configured channel (Placet, Telegram, Discord, custom), with native file attachments and conversation continuity. Here's how the delivery architecture works and why media-aware messaging matters for production agents.

Product

Facio's exec Tool: How AI Agents Run Shell Commands Safely With Timeouts and Audit Trails

Jun 10, 2026Product

Facio's exec Tool: How AI Agents Run Shell Commands Safely With Timeouts and Audit Trails

AI agents that can't execute code are research tools, not workers. Facio's exec tool gives agents controlled access to a real shell — with configurable timeouts, output truncation, dangerous-command blocking, and full audit trails. Here's how the safety architecture works and why it matters for production agents.

Product

Facio's Web Research Stack: How web_search and web_fetch Turn AI Agents Into Live Research Engines

Jun 9, 2026Product

Facio's Web Research Stack: How web_search and web_fetch Turn AI Agents Into Live Research Engines

AI agents with training data cutoffs are answering questions about a world that has moved on. Facio's web_search and web_fetch tools give agents direct access to the live web — search for sources, fetch the actual content, extract what's relevant, and reason over it. Combined with the search-reason loop, agents can run production-grade research workflows that match what humans do with browser tabs.

Product

Facio's Notebook Editing: How AI Agents Read and Modify Jupyter Notebooks at the Cell Level

Jun 8, 2026Product

Facio's Notebook Editing: How AI Agents Read and Modify Jupyter Notebooks at the Cell Level

Jupyter notebooks are a mess for AI agents. The JSON structure obscures the actual code, cells are interleaved with outputs and metadata, and most agents treat the whole notebook as opaque text. Facio's notebook_edit tool works at the cell level — read, replace, insert, and delete individual code and markdown cells with surgical precision. Here's why notebook-aware editing matters for data science agents.