--- schema_version: "1" id: 85a2ab3bba12223d2c53a372d32ebaa08c1fad1cb1461e464e3a5c05f2fd6a05 title: Instruction Adherence collection: admin version: current locale: en-us portal_url: https://docs.salesforce.com/admin/current/en-us/ai/264-0-0/agent_guardrails_instruction_adherence.md canonical_url: https://help.salesforce.com/s/articleView?id=ai.agent_guardrails_instruction_adherence.htm&release=264.0.0&type=5 source_site: help.salesforce.com content_type: documentation content_hash: 0d2c4d64d5dff90ad5c0edb390d5a57167f54d6125042719b60ceacd3d44ed9e description: Instruction adherence is an agent guardrail that evaluates how well your agents interpret subagent instructions when generating responses. products: - Agentforce - Cross Cloud release: "264" --- # Instruction Adherence Instruction adherence is an agent guardrail that evaluates how well your agents interpret subagent instructions when generating responses. ## Instruction Adherence Score Instruction adherence evaluates the agent's response and scores it based on how well the agent interprets and follows the subagent instructions. There are three score categories. | Score | Description | | --- | --- | | High | Indicates that the agent interprets and fully follows the subagent instructions, addressing key points and providing the required information. | | Low | Indicates that the agent doesn’t interpret or follow the subagent instructions accurately. It fails to follow at least one instruction, leading to an incorrect response. | | Uncertain | Indicates that instruction adherence can't be conclusively determined. This can be due to an ambiguous response or action, incomplete response, or conflicting interpretations of the subagent instructions. | ## Scores and Explanations Are Logged in Data 360 and Visible in Agent Event Logs Instruction Adherence scores are logged in the Einstein Generative AI Audit and Feedback data model in Data 360, which must be turned on in your org (see [Set Up Einstein Generative AI Audit and Feedback](https://help.salesforce.com/s/articleView?id=ai.generative_ai_feedback_enable.htm&release=264.0.0&type=5 "To collect Einstein generative AI audit and feedback data, turn it on in Einstein Setup. For orgs with only one dataspace, it’s turned on by default when Agentforce is turned on.")). Scores are visible in agent event logs. Turn on [Enhanced Agent Event Logs](https://help.salesforce.com/s/articleView?id=ai.copilot_setup_enhanced_event_logs.htm&release=264.0.0&type=5 "Event logs capture the events and user messages in an agent session to help you test and troubleshoot your agent. In supported agent types, enable enhanced event logs to store conversation data and view all session activity in one place. When enhanced event logs aren't enabled, events associated with agent and user messages (such as Message sent: and User response:) are logged, but the messages themselves aren't included in the event."). In addition to the scores, an explanation of why an agent's response was classified as high, low, or uncertain is included. You can use this information to see how your agent is following the subagent instructions. It can also help you determine whether the subagent instructions need improvement or more instructions. See [Best Practices for Writing Subagent Instructions and Troubleshooting Agents](https://help.salesforce.com/s/articleView?id=ai.copilot_topics_instructions.htm&release=264.0.0&type=5 "Subagent instructions set guidelines for agent behavior and give an agent the context it needs to do its job. Review best practices for writing subagent instructions to reduce latency and help your agent respond accurately and consistently."). ![Note](https://sf-zdocs-cdn-prod.zoominsoftware.com/tdta-ai-generative_ai-264-0-0-production-enus/f8ef6498-d7b2-44e3-9b34-ffd5a850aae9/images/icon_note.png) Note This feature is supported only in English and is available only for Agentforce for Service agents. ## Visualize Instruction Adherence in Agent Analytics Agent Analytics shows you the instruction adherence rate for a Service Agent under the Trust tab. See [Get Insights with Agent Analytics](https://help.salesforce.com/s/articleView?id=ai.generative_ai_agent_analytics_use.htm&release=264.0.0&type=5 "Monitor agent performance in Agentforce Studio with Agent Analytics dashboards, then open Sessions & Intents to audit transcripts, compare intents, and address low-quality sessions."). ![Instruction Adherence Rate in Agent Analytics](https://sf-zdocs-cdn-prod.zoominsoftware.com/tdta-ai-generative_ai-264-0-0-production-enus/f8ef6498-d7b2-44e3-9b34-ffd5a850aae9/generative_ai/images/agent_guardrails_instruction_adherence.png) - **[Create Data 360 Reports to Monitor Instruction Adherence](https://help.salesforce.com/s/articleView?id=ai.agent_instruction_adherence_monitor_dc.htm&release=264.0.0&type=5)** Create Data 360 reports to monitor how well agents are interpreting subagent instructions. #### See Also - [Create Data 360 Reports to Monitor Instruction Adherence](https://help.salesforce.com/s/articleView?id=ai.agent_instruction_adherence_monitor_dc.htm&release=264.0.0&type=5 "Create Data 360 reports to monitor how well agents are interpreting subagent instructions.") - _Agentforce Developer Guide_: Use Test Results to Improve Your Agent - [Task Resolution](https://help.salesforce.com/s/articleView?id=ai.agent_guardrails_task_resolution.htm&release=264.0.0&type=5 "Task resolution is an agent guardrail. It evaluates how well your agents interpret the user's intent, as expressed in a given conversation, and how well it resolves the task.")