---
title: "Creatio AI Studio adds memory-search planning and observability"
slug: "creatio-ai-studio-adds-memory-search-planning-and-observability"
published: "2026-10-16"
beat: "Launches"
tags: ["Launches", "Tools"]
creator: "Agentry Newsroom"
editor: "Susanne Sperling, Editor — Human in the Loop"
tools: ["Claude (Anthropic)", "Perplexity Sonar"]
creativeWorkStatus: "verified"
dateReviewed: "2026-10-16"
aiActArticle50: "compliant"
humanView: "https://agentry.news/launches/creatio-ai-studio-adds-memory-search-planning-and-observability"
agentView: "https://agentry.news/agent/creatio-ai-studio-adds-memory-search-planning-and-observability"
---# Creatio AI Studio adds memory-search planning and observability

> Creatio released updates to its AI Studio in September 2026 that add configurable model selection for memory-search planning and step-level visibility into agent memory access, along with new permissi

*Drafted by an AI agent. Verified by Susanne Sperling, Editor — Human in the Loop. [AI policy](/ai-policy).*

Creatio rolled out two updates to its AI Studio between September 14 and September 23, 2026, expanding observability and control over how agentic workflows access and plan memory searches.

## Configurable Memory-Search Planning

On September 14, 2026, Creatio added configurable model selection for memory-search planning in AI Studio [Agentry News](https://agentry.news/agent/creatio-adds-memory-search-planning-and-observability-to-ai-studio). The feature allows builders to specify which language model should handle the planning logic when agents decide whether and how to retrieve stored context or historical data. This shift from a fixed model to a configurable one gives teams flexibility to optimize cost, latency, and reasoning quality based on their specific workload—critical for enterprises running high-volume agent workflows where memory retrieval decisions compound across thousands of interactions.

## Step-Level Memory Visibility

The same September 14 release added step-level memory visibility, enabling builders and operators to see exactly when and how an agent accesses memory during execution [Agentry News](https://agentry.news/launches/creatio-adds-memory-search-planning-and-observability-to-ai-studio). Previously, memory operations were often opaque—visible only at a coarse level or buried in logs. Step-level granularity means that within a single agent run, each discrete action or decision point now surfaces its memory interactions, improving debugging, auditing, and compliance capabilities.

## New Permission Controls for Observability Data

On September 23, 2026, Creatio added a new **"Read run step content"** permission to control access to observability data [Agentry News](https://agentry.news/launches/creatio-adds-memory-search-planning-and-observability-to-ai-studio). This permission-level control allows enterprises to grant observability access to specific roles—such as compliance officers, security teams, or designated operators—without exposing sensitive run details to all users. The granular permission model reflects the growing need for audit trails and data governance in regulated industries where agents handle financial data, customer records, or proprietary information.

## Why It Matters for Agent Operations

These three changes address a core tension in agent deployment: **transparency vs. complexity**. As agents make more autonomous decisions about memory retrieval and planning, operators need visibility into what those agents see and how they reason. Configurable model selection lets teams tune that reasoning for their economics; step-level visibility makes debugging and auditing tractable; and permission controls ensure sensitive observability data stays compartmentalized. Together, they signal Creatio's focus on the operational layer of agent systems—the tooling and oversight mechanisms that separate production-ready agent platforms from experimental frameworks.

For enterprises evaluating low-code agent builders, observability depth and configurability are increasingly table-stakes differentiators.