skills.rest
padjei
padjeiCOMMUNITY
Explore all skills by this author →
GitHub

agentforce-observe

Analyze Data Cloud session traces to improve Agentforce agent performance.

Updated Jul 2, 2026
Search Tags:#agent optimization#agentforce#data cloud#stdm#session trace
One-click install
npx skills add https://github.com/padjei/SF_Agents --skill agentforce-observe-padjei
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agentforce-observe
Source: https://github.com/padjei/SF_Agents/tree/main/.claude/skills/agentforce-observe
Command: npx skills add https://github.com/padjei/SF_Agents --skill agentforce-observe-padjei

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires sf, sfcli, apex, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill addresses the challenge of analyzing and improving the performance of Agentforce agents by leveraging session traces and Data Cloud data, enabling users to identify and address issues such as misrouted subagents, missing actions, knowledge gaps, and adherence to instructions.

Core Features & Use Cases

  • Session Trace Analysis: Query and analyze session traces from Data Cloud to gain insights into agent behavior.
  • Reproduction of Issues: Simulate problematic conversations using local traces for reproduction and investigation.
  • Optimization: Make direct edits to the .agent file to improve agent configuration and performance.
  • Use Case: A user suspects that an Agentforce agent is misrouting conversations to the wrong subagent. This Skill allows them to analyze session traces, reproduce the issue, and update the .agent file to correct the routing.

Quick Start

Use the agentforce-observe skill to analyze the behavior of an Agentforce agent using session traces from the Data Cloud. Provide the org alias and any relevant details about the agent or session.

Frequently Asked Questions about agentforce-observe

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I analyze Agentforce session traces to find misrouted subagents?▼

You can analyze Agentforce session traces by querying Data Cloud STDM objects to identify misrouted subagents, missing actions, and knowledge gaps. This Skill extracts local traces to reproduce problematic conversations and investigate agent behavior.

What is the best way to optimize an Agentforce agent using session trace data?▼

Optimizing Agentforce agents involves analyzing Data Cloud session traces to pinpoint instruction adherence issues. After identifying knowledge gaps, you make direct edits to the `.agent` file to correct routing and improve the overall agent configuration.

Do I need Data Cloud STDM data model objects to analyze Agentforce behavior?▼

Yes, you need STDM Data Model Objects in Data Cloud to query and analyze Agentforce behavior. The Skill relies on these objects to extract session traces and requires knowledge of Agentforce metadata to function correctly.

Can I edit the .agent file directly to fix Agentforce knowledge gaps?▼

Yes, you can edit the `.agent` file directly to fix Agentforce knowledge gaps and configuration issues. This Skill supports direct metadata editing after analyzing local traces and reproducing problematic conversations.

How do I deploy Agentforce metadata changes using sf CLI?▼

To deploy Agentforce metadata changes, this Skill utilizes the sf CLI and requires Apex classes for STDM querying. You provide the org alias and relevant session details to execute the deployment of the updated `.agent` file.

Related Skills

Explore More Software Engineering Skills→
cchrisvcchrisv

observing-agentforce

Analyze production Agentforce agent behavior using STDM session traces and Data Cloud records.

Community
Advanced
forcedotcomforcedotcom
803

testing-agentforce

Verify Agentforce agent behavior with smoke and regression tests via sf CLI.

Official
Advanced
padjeipadjei

platform-metadata-deploy

Automate Salesforce metadata deployment and orchestration using sf CLI v2.

Community
Advanced
dmckeeSFdmckeeSF

observing-agentforce

Analyze production Agentforce agent behavior using session traces and Data Cloud.

Community
Advanced