Facio Blog
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 146-150 of 235 articles.
Engineering
The official Filesystem MCP server from the Model Context Protocol team — 14 tools, 86K+ downloads, full Tool Annotations, directory access control via Roots, and the safety patterns every other server follows.
Product
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.
Security
Microsoft added "AI" as a fifth pillar to its Zero Trust Workshop with 700 controls. Anthropic published a three-tier framework the same week. CISA's ATF from CSA, the NSA's MCP guidance, and the OWASP Top 10 all point to the same answer. The convergence is the story.
Engineering
HasData exposes 40+ site-specific MCP tools — Google Search, Maps, Trends, Flights, Amazon, Zillow, Airbnb, Indeed, Glassdoor, Yelp — returning parsed JSON instead of raw HTML. Hosted Streamable HTTP transport, one API key, no local process.
Product
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.