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 46-50 of 235 articles.

Engineering

MCP Spotlight: Brave Search MCP Server — The Web-Wide Search Bridge With Independent Index, Local-First Privacy Options, and the Research-Default Reference

Jul 12, 2026Engineering

MCP Spotlight: Brave Search MCP Server — The Web-Wide Search Bridge With Independent Index, Local-First Privacy Options, and the Research-Default Reference

The official Brave Search MCP Server by Brave Software — ~8 focused tools covering web, local, image, video, and news search. Brave's independent web crawler (no Google/Bing dependency), privacy-first (no user tracking), 2000 queries/month free tier, EU presence. The independent-index search default for AI agents in 2026.

Product

Facio's Backpressure Discipline: How AI Agent Systems Slow Down Before They Break

Jul 12, 2026Product

Facio's Backpressure Discipline: How AI Agent Systems Slow Down Before They Break

Production AI agent systems face the same problem as any distributed system: capacity limits. The agents can process N tasks per minute; the upstream systems can accept M requests per second; downstream services can handle K concurrent calls. When input exceeds capacity, naive systems keep accepting work and pile it up. The pile grows until memory exhausts, queues overflow, or the system crashes. Facio's backpressure discipline gives systems the structural answer: detect overload, signal backpressure, shed load gracefully, recover when capacity returns.

Human-in-the-loop

HITL and the Cost of Saying Maybe: Why Escalation Is the Most Underrated Decision in Human Oversight

Jul 12, 2026Human-in-the-loop

HITL and the Cost of Saying Maybe: Why Escalation Is the Most Underrated Decision in Human Oversight

Most HITL designs treat escalation as a fallback — what happens when the reviewer doesn't know. But escalation is a first-class decision with its own costs, its own value, and its own failure modes. A reviewer who escalates appropriately is doing the most underrated work in HITL. Here's why "I don't know, escalate" is the third decision that deserves the same support as approve and reject.

Engineering

MCP Security Architecture 2026: Threat Models, Sandboxing, and the Defense-in-Depth Layers Every Production Agent Deployment Needs

Jul 11, 2026Engineering

MCP Security Architecture 2026: Threat Models, Sandboxing, and the Defense-in-Depth Layers Every Production Agent Deployment Needs

The defense-in-depth reference architecture every team deploying MCP in production needs. Six layers: server verification → process sandboxing → credential isolation → per-tool RBAC → HITL approval → audit trail. Plus prompt-injection defense and multi-agent trust boundaries. Companion to the MCP Spotlight series.

Product

Facio's Cascading Timeouts: How AI Agents Bound Latency at Every Layer of the Stack

Jul 11, 2026Product

Facio's Cascading Timeouts: How AI Agents Bound Latency at Every Layer of the Stack

An AI agent's response time is the sum of many operations: model inference, tool calls, database queries, API requests, HITL pauses. Each operation has its own latency. Each operation can fail by being slow. Without discipline, a single slow operation makes the whole agent slow. Facio's cascading timeouts give teams the structural discipline to bound latency at every layer of the stack. The agent has time budget per operation; the operations chain into a budget per session; the session fails fast rather than hanging.