Agentic Index: agent builders change log https://agenticindex.io/rankings/agent-builders Verified changes to agent builders, from the Agentic Index change log. Every entry is dated and sourced from public vendor material. en-us Free to reuse with a visible link to agenticindex.io. Terms: https://agenticindex.io/use-this-data Sat, 26 Sep 2026 12:00:00 GMT 720 https://media.base44.com/images/public/69e4cec10b4126d233ee7ee4/a44c0d34a_generated_image.png Agentic Index: agent builders change log https://agenticindex.io/rankings/agent-builders Boomi Agent Control Tower now supports adding provider agents at the account level, manually or through AI Catalog. https://agenticindex.io/change-log/boomi agenticindex.io:change:boomi-sfjuge-2026-09-26 Sat, 26 Sep 2026 12:00:00 GMT Observability / auditability Enterprise operations agent Agent builder Multi-agent platform Boomi Agent Control Tower now supports adding provider agents at the account level, manually or through AI Catalog. When connecting Salesforce, Microsoft Copilot, or Snowflake Cortex, customers can choose scheduled Boomi Sync or API Push. API Push lets customers supply metadata and usage metrics that those providers do not return through Boomi Sync.

Why it matters: Organizations can choose between automatic inventory synchronization and supplying the measurements needed for agent oversight. API Push requires customers to implement and maintain the data feed, so monitoring coverage depends on their integration.

Impact: Medium. Verification: Partially Verified. Type: Observability / auditability.

Evidence: Official changelog

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Replit Agent can now analyze an uploaded dataset or connected data source and display interactive charts directly in chat, using… https://agenticindex.io/change-log/replit agenticindex.io:change:replit-3iddu3-2026-09-25 Fri, 25 Sep 2026 12:00:00 GMT Agent capability Agent builder Replit Agent can now analyze an uploaded dataset or connected data source and display interactive charts directly in chat, using technology from its acquisition of Atta. Users can request a specific chart or ask the agent to visualize findings, with the agent handling queries and analytical steps.

Why it matters: Business teams can investigate data and inspect visual results within the same conversation, reducing the need to hand each follow up question to a separate analyst or charting tool.

Impact: Medium. Verification: Verified. Type: Agent capability.

Evidence: Official changelog

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Windmill v1.818.0 counts enabled service accounts as operator seats in self hosted Enterprise deployments, billing each as half a seat… https://agenticindex.io/change-log/windmill agenticindex.io:change:windmill-170vtl-2026-09-25 Fri, 25 Sep 2026 12:00:00 GMT Pricing / packaging Agent infrastructure Agent builder Windmill v1.818.0 counts enabled service accounts as operator seats in self hosted Enterprise deployments, billing each as half a seat regardless of its assigned role. Enabled service accounts also count toward the offline license seat cap, and creation is refused when the cap is reached. The related Cloud Team billing change is separate from this release.

Why it matters: Organizations using dedicated identities for agents and automation need to include those accounts in license forecasts. Reviewing unused enabled accounts before upgrading can help avoid unexpected charges or blocked account creation.

Impact: High. Verification: Verified. Type: Pricing / packaging.

Evidence: GitHub release

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Box introduced Doc Split APIs to divide a PDF into separate documents and save them to a chosen Box folder. https://agenticindex.io/change-log/box agenticindex.io:change:box-18gg8z-2026-09-24 Thu, 24 Sep 2026 12:00:00 GMT MCP / tool calling / API Enterprise operations agent Agent builder Box introduced Doc Split APIs to divide a PDF into separate documents and save them to a chosen Box folder. Manual mode uses explicit page ranges, while smart mode interprets natural language instructions such as splitting at QR code separators or isolating signature pages. The asynchronous API creates a job through POST /2.0/document_splits and exposes its status and generated files through a GET endpoint. Requests use API version 2026.0, and smart splits consume one AI Unit per four pages.

Why it matters: Teams can automate document packet preparation before extraction or review without manually separating PDFs. Buyers should account for AI Unit consumption and verify the returned page groupings before routing sensitive documents.

Impact: High. Verification: Verified. Type: MCP / tool calling / API.

Evidence: Official changelog

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Tines released Records API v2 with endpoints for reading records and record types, searching records, and aggregating results. https://agenticindex.io/change-log/tines agenticindex.io:change:tines-xug3l0-2026-09-24 Thu, 24 Sep 2026 12:00:00 GMT MCP / tool calling / API Security / SOC agent Agent builder Enterprise operations agent Tines released Records API v2 with endpoints for reading records and record types, searching records, and aggregating results. Requests can select fields by name or ID and include linked cases, child records, and full artifacts in the same call. The API standardizes response and error handling for workflows that retrieve Records data.

Why it matters: Teams building agent workflows against Tines can reduce chained retrieval calls and custom parsing logic. The new version provides a clearer integration surface for retrieving only the operational context a workflow needs.

Impact: Medium. Verification: Verified. Type: MCP / tool calling / API.

Evidence: Official changelog

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Voiceflow made live agent handoff available on every plan. https://agenticindex.io/change-log/voiceflow agenticindex.io:change:voiceflow-130hl0-2026-09-24 Thu, 24 Sep 2026 12:00:00 GMT Pricing / packaging Customer support agent Agent builder Voice agent Voiceflow made live agent handoff available on every plan. Chat agents can use the Handoff step to route conversations to Ujet, Genesys, Kustomer, Dixa, Sunshine Conversations for Zendesk, or a custom handoff provider.

Why it matters: Teams on lower plans can include human escalation in a chat deployment without changing their Voiceflow plan to unlock that step. Buyers can assess how the handoff fits their existing contact center and support routing.

Impact: High. Verification: Partially Verified. Type: Pricing / packaging.

Evidence: Official changelog

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Activepieces 0.92.0 adds agent creation from a natural language description and an AI Router that chooses a workflow route from a question. https://agenticindex.io/change-log/activepieces agenticindex.io:change:activepieces-lan9qs-2026-09-24 Thu, 24 Sep 2026 12:00:00 GMT Agent capability Agent builder Activepieces 0.92.0 adds agent creation from a natural language description and an AI Router that chooses a workflow route from a question. The release also adds a Deno code sandbox, an MCP activity feed recording client calls, and OpenID Connect support for restricting connector access by email domain.

Why it matters: Teams can build agents with less manual setup and inspect the actions external MCP clients execute. Domain restrictions give administrators an additional control when exposing connected business systems.

Impact: Medium. Verification: Verified. Type: Agent capability.

Evidence: GitHub release

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Base44 moved Apps API and white label integration authentication to personal access tokens and published an October 15, 2026 retirement… https://agenticindex.io/change-log/base44 agenticindex.io:change:base44-1hwysp-2026-09-23 Wed, 23 Sep 2026 12:00:00 GMT MCP / tool calling / API Agent builder Base44 moved Apps API and white label integration authentication to personal access tokens and published an October 15, 2026 retirement date for account API keys. Integrations must replace the api_key header with Authorization: Bearer and a personal access token. Monitoring and Audit Logs APIs continue using workspace API keys.

Why it matters: Teams running scripts or integrations against the Apps API need to migrate credentials before the cutoff to preserve service continuity.

Impact: High. Verification: Partially Verified. Type: MCP / tool calling / API.

Evidence: Official changelog

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Agent Zero v2.13 replaces runtime=input with input_remote for remote execution, without a compatibility alias. https://agenticindex.io/change-log/agent-zero agenticindex.io:change:agent-zero-1ci28w-2026-09-23 Wed, 23 Sep 2026 12:00:00 GMT MCP / tool calling / API Multi-agent platform Browser / computer-use agent Coding agent Agent builder Agent Zero v2.13 replaces runtime=input with input_remote for remote execution, without a compatibility alias. It also adds Telegram and WhatsApp slash commands with project aware command catalogs and scoped integration settings, plus a configurable browser JavaScript evaluation timeout.

Why it matters: Existing remote execution workflows need a migration before upgrading. Operators gain more direct control over agents through messaging channels and a recovery mechanism for browser scripts that stall.

Impact: High. Verification: Verified. Type: MCP / tool calling / API.

Evidence: GitHub release

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> FlowX.AI documented a Browser Automation node for Agent Builder, available from release 5.13.0 on SaaS. https://agenticindex.io/change-log/flowx-ai agenticindex.io:change:flowx-ai-b7s3kq-2026-09-23 Wed, 23 Sep 2026 12:00:00 GMT Agent capability Agent builder Enterprise operations agent Agent infrastructure Multi-agent platform FlowX.AI documented a Browser Automation node for Agent Builder, available from release 5.13.0 on SaaS. The node drives a real browser through web applications that have no API, filling multi page forms, navigating portals and retrieving results from a task written in plain language. It checks outcomes against the target site's recorded state rather than the agent's own report and keeps the browser session across workflow steps. Self hosted deployments receive it with the next LTS release family.

Why it matters: Banks and insurers running agents on FlowX can now reach legacy portals and third party sites that expose no integration point from inside the same governed workflow, with every run logged. Verification against the site's recorded state is the detail to test in a pilot, since an agent reporting success on a form it never submitted is the usual failure in this category.

Impact: High. Verification: Verified. Type: Agent capability.

Evidence: Official documentation

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> AutoGPT Platform v0.8.0 generates secrets per installation, refuses the previously published default encryption key, and removes legacy… https://agenticindex.io/change-log/autogpt agenticindex.io:change:autogpt-g25165-2026-09-19 Sat, 19 Sep 2026 12:00:00 GMT Security / enterprise Agent builder Agent infrastructure Multi-agent platform AutoGPT Platform v0.8.0 generates secrets per installation, refuses the previously published default encryption key, and removes legacy HS256 JWT verification. Existing installations using the defaults must migrate stored integration credentials to a new key. The release also adds scheduled expert routines, spending threshold approvals, and restrictions that let experts run only installed workflows with granted credentials.

Why it matters: Self hosted operators need to plan the credential migration before upgrading to avoid startup failures or losing access to connected services. Teams delegating recurring work gain clearer controls over cost and authorized actions.

Impact: High. Verification: Verified. Type: Security / enterprise.

Evidence: GitHub release

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Swimlane released Turbine 26.4, which adds structured JSON schema outputs from Hero AI actions, custom dynamic webhook responses and… https://agenticindex.io/change-log/swimlane agenticindex.io:change:swimlane-1ocrhev-2026-09-17 Thu, 17 Sep 2026 12:00:00 GMT Agent capability Security / SOC agent Agent builder Swimlane released Turbine 26.4, which adds structured JSON schema outputs from Hero AI actions, custom dynamic webhook responses and replay for package imports.

Why it matters: Structured outputs make AI actions safe to feed into the next automation step. Free text outputs are where security playbooks break.

Impact: Medium. Verification: Partially Verified. Type: Agent capability.

Evidence: Official announcement

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Celonis added Cloud Functions to its Orchestration Engine, so builders can call third party services or AI agents, and expanded its Agent… https://agenticindex.io/change-log/celonis agenticindex.io:change:celonis-u5cmwe-2026-09-16 Wed, 16 Sep 2026 12:00:00 GMT Workflow orchestration Enterprise operations agent Agent builder Agent infrastructure Celonis added Cloud Functions to its Orchestration Engine, so builders can call third party services or AI agents, and expanded its Agent Toolkit through MCP.

Why it matters: Process mining data is most valuable when agents can act on it. Calling external agents from Celonis turns analysis into automation, with the governance questions that come with it.

Impact: Medium. Verification: Partially Verified. Type: Workflow orchestration.

Evidence: Official blog

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Dataiku launched Cobuild on Snowflake, which combines Snowflake Cortex AI's model access with Dataiku's orchestration layer so users can… https://agenticindex.io/change-log/dataiku agenticindex.io:change:dataiku-1ixx6rp-2026-09-15 Tue, 15 Sep 2026 12:00:00 GMT Integrations Multi-agent platform Agent builder Data analyst agent Dataiku launched Cobuild on Snowflake, which combines Snowflake Cortex AI's model access with Dataiku's orchestration layer so users can build and validate AI workflows and agents in plain language.

Why it matters: Building agents where the data already lives avoids copying governed data into another tool. Snowflake customers should check how Dataiku's governance and Snowflake's access controls combine.

Impact: High. Verification: Partially Verified. Type: Integrations.

Evidence: News article

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Airrived launched Agentic Observability in its Agentic OS, tracing agent behavior end to end, from the integration an agent uses to the… https://agenticindex.io/change-log/airrived agenticindex.io:change:airrived-63ssxn-2026-09-14 Mon, 14 Sep 2026 12:00:00 GMT Observability / auditability Security / SOC agent Agent builder Enterprise operations agent Airrived launched Agentic Observability in its Agentic OS, tracing agent behavior end to end, from the integration an agent uses to the outcome it produces, including who created it and what permissions it holds.

Why it matters: Observability that starts at the permission and ends at the outcome answers the audit question directly: which agent did this, and was it allowed to.

Impact: High. Verification: Verified. Type: Observability / auditability.

Evidence: Press release

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Base44 released Superagent and AI chat upgrades, including GPT-6 Astra for reasoning. https://agenticindex.io/change-log/base44 agenticindex.io:change:base44-11y0gqp-2026-09-14 Mon, 14 Sep 2026 12:00:00 GMT Agent capability Agent builder Base44 released Superagent and AI chat upgrades, including GPT-6 Astra for reasoning. Agents can now execute record changes without a secondary confirmation step.

Why it matters: Removing the confirmation step is a real shift in autonomy for an app builder whose agents write to live data. Builders should decide deliberately which apps get it, rather than inherit it.

Impact: High. Verification: Partially Verified. Type: Agent capability.

Evidence: Official changelog

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Box added admin model controls for the Box Agent, giving administrators more governance over which models the agent uses and how it… https://agenticindex.io/change-log/box agenticindex.io:change:box-1zfecg-2026-09-14 Mon, 14 Sep 2026 12:00:00 GMT Security / enterprise Enterprise operations agent Agent builder Box added admin model controls for the Box Agent, giving administrators more governance over which models the agent uses and how it operates in their organization.

Why it matters: Model choice is a compliance decision for regulated content. Admin level control means the choice belongs to IT, not to each user.

Impact: Medium. Verification: Partially Verified. Type: Security / enterprise.

Evidence: Official documentation

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> StackBlitz Bolt: StackBlitz introduced Bolt Forge, a Bolt agent powered by open source models. https://agenticindex.io/change-log/bolt agenticindex.io:change:bolt-1wxir95-2026-09-14 Mon, 14 Sep 2026 12:00:00 GMT Agent capability Agent builder StackBlitz introduced Bolt Forge, a Bolt agent powered by open source models. It is a research preview that gives individual Pro users up to 50 times more usage at no extra cost.

Why it matters: Open source models are how Bolt can offer 50 times the usage at the same price. The trade to test is quality on real projects, since this is a research preview.

Impact: High. Verification: Verified. Type: Agent capability.

Evidence: Official blog

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Serval announced the general availability of Catalyst, an AI agent designed to build enterprise automations. https://agenticindex.io/change-log/serval agenticindex.io:change:serval-6z3ybd-2026-09-04 Fri, 04 Sep 2026 12:00:00 GMT Agent capability Enterprise operations agent Agent builder Serval announced the general availability of Catalyst, an AI agent designed to build enterprise automations. Catalyst is now enabled by default for all organizations, allowing them to instantly generate automations from ticket history, natural language instructions, or existing standard operating procedures.

Why it matters: This significantly accelerates time-to-value by automating the creation of workflows rather than requiring manual configuration. Buyers can leverage their existing historical data and documentation to immediately deploy functional automations across their enterprise.

Impact: High. Verification: Verified. Type: Agent capability.

Evidence: Official announcement

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Base44 introduced a Model Context Protocol (MCP) server and sandbox commands that allow developers to connect their own external AI coding… https://agenticindex.io/change-log/base44 agenticindex.io:change:base44-ciwuyf-2026-09-03 Thu, 03 Sep 2026 12:00:00 GMT MCP / tool calling / API Agent builder Base44 introduced a Model Context Protocol (MCP) server and sandbox commands that allow developers to connect their own external AI coding agents to a Base44 app's cloud environment. The integration supports listing, reading, searching, and editing files directly via the CLI without a local checkout.

Why it matters: This significantly expands extensibility for advanced engineering teams, allowing them to leverage their preferred external AI assistants and custom scripts to modify Base44 applications programmatically.

Impact: High. Verification: Partially Verified. Type: MCP / tool calling / API.

Evidence: Official changelog

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Relevance AI rolled out a multi-feature update introducing the Invent interface for agent creation and adding Google's Gemini 3.5 Flash as… https://agenticindex.io/change-log/relevance-ai agenticindex.io:change:relevance-ai-15p7c98-2026-09-03 Thu, 03 Sep 2026 12:00:00 GMT Agent capability Multi-agent platform Agent builder Relevance AI rolled out a multi-feature update introducing the Invent interface for agent creation and adding Google's Gemini 3.5 Flash as a selectable language model across all workflows. The release also updated the platform's MCP server integration to enforce user-level authentication for agent access.

Why it matters: The addition of Gemini 3.5 Flash allows users to process large 1M-token contexts and multimodal inputs while tuning reasoning depth to optimize for speed or cost. The MCP authentication changes also provide strict access control by ensuring users can only trigger agents they explicitly have permission to access.

Impact: High. Verification: Partially Verified. Type: Agent capability.

Evidence: Official changelog

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Boomi launched the Agent Control Plane and Boomi AI Gateway to provide centralized governance and visibility for enterprise AI agents. https://agenticindex.io/change-log/boomi agenticindex.io:change:boomi-k6otqn-2026-09-02 Wed, 02 Sep 2026 12:00:00 GMT Security / enterprise Enterprise operations agent Agent builder Multi-agent platform Boomi launched the Agent Control Plane and Boomi AI Gateway to provide centralized governance and visibility for enterprise AI agents. The new infrastructure combines MCP and LLM gateway capabilities to inspect live traffic, apply identity and rate limits, and hold high-risk transactional actions for human approval.

Why it matters: IT and security teams can safely deploy AI agents across multi-cloud environments without losing control over data or costs. This mitigates risks related to token overruns, compliance exposure, and unauthorized system access by enforcing strict policies.

Impact: High. Verification: Partially Verified. Type: Security / enterprise.

Evidence: Press release

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Base44: Enterprise workspaces can now create service principals to provide automated systems with their own distinct identities. https://agenticindex.io/change-log/base44 agenticindex.io:change:base44-11y0gpp-2026-09-01 Tue, 01 Sep 2026 12:00:00 GMT Security / enterprise Agent builder Enterprise workspaces can now create service principals to provide automated systems with their own distinct identities. Administrators can issue and revoke credentials specifically for these non-human accounts.

Why it matters: This is a critical security and governance feature for enterprise buyers, ensuring that automated workflows and API integrations do not rely on individual user accounts that could be compromised or deactivated.

Impact: High. Verification: Partially Verified. Type: Security / enterprise.

Evidence: Official changelog

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Serval launched scheduled workflows, enabling users to run automations on any defined cadence. https://agenticindex.io/change-log/serval agenticindex.io:change:serval-7syuek-2026-08-31 Mon, 31 Aug 2026 12:00:00 GMT Workflow orchestration Enterprise operations agent Agent builder Serval launched scheduled workflows, enabling users to run automations on any defined cadence. The feature supports time-based executions such as daily ticket summaries, weekly reports, and hourly SLA alerts.

Why it matters: This allows IT and operations teams to move beyond purely reactive, trigger-based automations to proactive, time-based operational tasks. It reduces manual reporting overhead and ensures critical alerts or summaries are delivered consistently without human intervention.

Impact: Medium. Verification: Verified. Type: Workflow orchestration.

Evidence: Official announcement

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Secrets and third-party integrations can now be managed programmatically via the stable API, CLI, and the Voiceflow MCP server. https://agenticindex.io/change-log/voiceflow agenticindex.io:change:voiceflow-cin2or-2026-08-31 Mon, 31 Aug 2026 12:00:00 GMT MCP / tool calling / API Customer support agent Agent builder Voice agent Secrets and third-party integrations can now be managed programmatically via the stable API, CLI, and the Voiceflow MCP server. Secret values are strictly write-only and are automatically refused if they arrive from an MCP client.

Why it matters: Engineering teams can fully automate the provisioning of agent environments and securely manage credentials in CI/CD pipelines without requiring manual intervention in the visual builder.

Impact: High. Verification: Partially Verified. Type: MCP / tool calling / API.

Evidence: Official changelog

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Dify launched New Agent, which allows agents to operate as standalone and independent applications. https://agenticindex.io/change-log/dify agenticindex.io:change:dify-t4p26o-2026-08-29 Sat, 29 Aug 2026 12:00:00 GMT Agent capability Agent builder Dify launched New Agent, which allows agents to operate as standalone and independent applications. These agents can also be utilized as reusable resources within larger workflows.

Why it matters: This architectural update provides teams with the flexibility to deploy agents independently or embed them modularly into complex business workflows, streamlining AI application development.

Impact: High. Verification: Verified. Type: Agent capability.

Evidence: Official blog

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Windmill enabled Anthropic prompt caching specifically for the Vertex AI agent. https://agenticindex.io/change-log/windmill agenticindex.io:change:windmill-360a31-2026-08-28 Fri, 28 Aug 2026 12:00:00 GMT Agent capability Agent infrastructure Agent builder Windmill enabled Anthropic prompt caching specifically for the Vertex AI agent. This update allows agents utilizing Anthropic models via Google Cloud's Vertex AI to cache context and instructions.

Why it matters: Prompt caching significantly reduces latency and API costs for workflows that reuse large contexts or system prompts. This makes deploying complex Anthropic-powered agents on Vertex AI much more economical for enterprise users.

Impact: Medium. Verification: Verified. Type: Agent capability.

Evidence: GitHub release

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Haptik introduced a Parameter Store to serve as a centralized variable store for customer identifiers, verification flags, and preferences. https://agenticindex.io/change-log agenticindex.io:change:haptik-1qkn69n-2026-08-27 Thu, 27 Aug 2026 12:00:00 GMT Memory / state Agent builder Haptik introduced a Parameter Store to serve as a centralized variable store for customer identifiers, verification flags, and preferences. The feature provides deterministic state control with real-time read and write capabilities, cross-channel session persistence, and Time-to-Live expiration rules.

Why it matters: Enterprises can maintain strict schema validity across complex conversations, eliminating repetitive questions and ensuring secure handling of sensitive customer data across web, mobile, and voice channels.

Impact: Medium. Verification: Verified. Type: Memory / state.

Evidence: Official blog

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> AutoGPT introduced expert-scoped sessions and identity context capabilities to its Copilot interface. https://agenticindex.io/change-log/autogpt agenticindex.io:change:autogpt-nxvsdi-2026-08-25 Tue, 25 Aug 2026 12:00:00 GMT Agent capability Agent builder Agent infrastructure Multi-agent platform AutoGPT introduced expert-scoped sessions and identity context capabilities to its Copilot interface. This update enables the system to maintain specialized personas and contextual identity throughout user interactions.

Why it matters: Buyers building complex or specialized agents will benefit from improved continuity and relevance, as the agent can now hold onto a specific expert identity and context across a session.

Impact: Medium. Verification: Partially Verified. Type: Agent capability.

Evidence: GitHub release

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]> Base44 added per-app exceptions for channel usage policies, allowing administrators to specify and record reasons for channel access. https://agenticindex.io/change-log/base44 agenticindex.io:change:base44-11xzp1e-2026-08-25 Tue, 25 Aug 2026 12:00:00 GMT Security / enterprise Agent builder Base44 added per-app exceptions for channel usage policies, allowing administrators to specify and record reasons for channel access. These policies are enforced on Base44's servers to prevent agents from being manipulated into bypassing them.

Why it matters: This provides enterprise administrators with granular governance over agent communication channels. Server-side enforcement ensures that prompt injection or agent hallucinations cannot override security policies.

Impact: Medium. Verification: Partially Verified. Type: Security / enterprise.

Evidence: Official changelog

Source: Agentic Index change log. Free to reuse with a link back. Terms.

]]>