Singapore – AppsFlyer has expanded its agentic AI suite with new Agent Hub capabilities and Model Context Protocol (MCP) tools, enabling marketing teams to investigate campaign data, automate workflows, and execute governed, human-approved actions.

Built to provide transparent decision-making behind every AI recommendation, the release introduces custom workflow automation and direct action execution within Agent Hub. Marketers can now build custom agents using plain-language descriptions that pull event-level data, perform root cause analysis, and deliver scheduled insights via Slack or email. 

The update also introduces the Apple Ads Share of Voice Agent, currently in closed beta, which estimates share of voice by keyword and geography, recommends bid adjustments, and executes changes directly in Apple Ads upon user approval.

AppsFlyer also introduced four new MCP capabilities—available via its built-in AI assistant—to connect core marketing data with external AI tools. These updates surface Protect360 fraud and validation block metrics alongside creative-level spend, reach, conversion, retention, and revenue cohorts to help teams spot winning or fatiguing assets faster.

Marketers can also convert plain-language campaign briefs into configured OneLinks pending approval and view partner attribution windows, postback rules, and event mappings to simplify setup troubleshooting.

Inna Weiner, VP Product at AppsFlyer, said, “Marketing teams are moving from manual campaign orchestration to agentic execution. With this release, AppsFlyer is giving marketers trusted, agentic tools that can analyze, recommend, and act on their data to deliver clear value for growth.”

Through this release, AppsFlyer aims to help marketing teams transition from manual campaign management to trusted, autonomous execution while maintaining human oversight.