Productivity

Adobe CX Enterprise Coworker

by Adobe Inc.

Overview

With Adobe CX Enterprise Coworker, marketers and business practitioners can use natural language to access insights, analyze customer experiences, and complete marketing tasks across Adobe CX Enterprise. Coworker helps marketers expedite and automate workflows, accelerate decision-making, and turn ideas into action without requiring deep product expertise. Built with enterprise governance and context, it delivers trusted assistance at scale. Language Support: AI Assistant is currently supported in English only. Non-English inputs may produce inconsistent or erroneous results. Personal Data: If you include personal data in your prompts, only add what's necessary and only if you have the right to use it.

Tools

ask_aep_agent

ChatGPT

Adobe Experience Platform AI Assistant. Sends a message to run a full agent turn (LLM and tools) for AEP tasks — exploring data, answering questions, and driving supported workflows. Pass episode_id from the previous response to continue a conversation; omit it to start a new one. A status="input_required" response means the agent needs more info: relay its response and reply with the same episode_id.

core_list_dataviews

ChatGPT

Lists the dataviews available to the current organisation, with their name and ID. Use this only when the user asks which dataviews exist, or before switching so you can pick a valid name to pass to set_dataview. Do NOT use it as a setup or discovery step for data questions (reports, metrics, dimensions) — those tools operate on the already-active dataview and never require listing dataviews first. Read-only — it does not change the active dataview.

core_list_orgs

ChatGPT

Lists the Adobe organisations the current user belongs to, with their name and ID. Use this only when the user asks which organisations they have access to, or before switching so you can pick a valid org ID to pass to switch_org. Do NOT use it as a setup or discovery step for data questions (segments, schemas, audiences, datasets, destinations) — those tools operate on the already-active org and never require listing organisations first. Read-only — it does not change the active organisation.

core_list_sandboxes

ChatGPT

Lists the sandboxes available to the current organisation, with their name and type. Use this only when the user asks which sandboxes exist, or when they name a sandbox to switch to so you can pick a valid name for set_sandbox. Do NOT use it as a setup or discovery step for data questions (segments, schemas, audiences, datasets, destinations) — those tools operate on the already-active sandbox and never require listing sandboxes first. Read-only — it does not change the active sandbox or create/delete sandboxes.

core_list_sub_orgs

ChatGPT

Lists the sub-orgs available to the current organisation, with their name and ID. Use this only when the user asks which sub-orgs exist, or before switching so you can pick a valid ID to pass to set_sub_org. Do NOT use it as a setup or discovery step for data questions — those tools operate on the already-active sub-org and never require listing sub-orgs first. Read-only — it does not change the active sub-org.

get_suspend_resolution

ChatGPT

Widget-internal — you must not call this; the inline suspend widget calls it for you on mount to recover a turn that was already answered and render its settled state. It only reads state and never resumes a turn, and calling it advances nothing (without kind it just returns a validation error). To continue a suspended turn, call ask_aep_agent with the user's reply and the same episode_id (see its input_required guidance).

Capabilities

WritesInteractive

Example Prompts

Click any prompt to copy it.

Links

Documentation ↗Privacy Policy ↗

App Stats

Tools

Prompts

ChatGPT

Platforms