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 171-175 of 281 articles.

Product

How Facio Picks the Right Tool for the Job: The Decision Framework Behind Tool Selection in AI Agents

Jun 19, 2026Product

How Facio Picks the Right Tool for the Job: The Decision Framework Behind Tool Selection in AI Agents

A Facio agent has 25+ native tools plus dynamic MCP extensions. Picking the wrong tool for the task is one of the most common failure modes — agents waste iterations, produce wrong results, or trigger unnecessary HITL gates because they reached for the wrong primitive. Facio's tool selection framework turns tool choice into a decision the agent reasons about explicitly. Here's how the framework works and why tool selection is a learnable skill.

Human-in-the-loop

Five HITL Scaling Inflection Points: The Architecture That Breaks at Every Order of Magnitude

Jun 19, 2026Human-in-the-loop

Five HITL Scaling Inflection Points: The Architecture That Breaks at Every Order of Magnitude

The HITL roadmap that worked for five agents breaks at fifteen. The patterns that worked for one team don't transfer to five. The approval gates that worked for one action type break at thirty. Here are the five scaling inflection points and how to design for the order you'll hit them.

Engineering

MCP Spotlight: ArcadeDB — The Built-In MCP Server for the Only Multi-Model DBMS That Speaks SQL, Cypher, Gremlin, GraphQL, MongoQL & Redis

Jun 18, 2026Engineering

MCP Spotlight: ArcadeDB — The Built-In MCP Server for the Only Multi-Model DBMS That Speaks SQL, Cypher, Gremlin, GraphQL, MongoQL & Redis

ArcadeDB ships the only multi-model DBMS with a first-party built-in MCP server — 7 data models (graph, document, key/value, search, time series, vector, geospatial), 6 query languages (SQL, Cypher, Gremlin, GraphQL, MongoDB QL, Redis), 70+ graph algorithms, Apache 2.0.

Product

Facio's Reflection Process: How Periodic Memory Curation Keeps AI Agents Honest at Scale

Jun 18, 2026Product

Facio's Reflection Process: How Periodic Memory Curation Keeps AI Agents Honest at Scale

Memory that grows without curation becomes noise. Facts accumulate, contradictions pile up, stale entries linger, and the agent's passive context fills with information that no longer matters. Facio's Reflection process is the periodic curator that keeps MEMORY.md honest — identifying staleness, resolving contradictions, consolidating patterns, and pruning noise. Here's how the curator works and why automated memory hygiene is a non-negotiable for production agents.

Human-in-the-loop

The Reviewer Doesn't Know What the Agent Knows: Closing the Information Asymmetry in HITL

Jun 18, 2026Human-in-the-loop

The Reviewer Doesn't Know What the Agent Knows: Closing the Information Asymmetry in HITL

Most HITL designs assume the human knows what's being asked of them. But the reviewer's mental model — what the agent is doing, why, with what risk — is often incomplete or wrong. The agent knows things the reviewer doesn't. The reviewer guesses. The guess becomes the decision. How to fix the information asymmetry.