docs: add AI fallback use case combining Label the incoming message and Open a ticket#645
docs: add AI fallback use case combining Label the incoming message and Open a ticket#645SangeetaMishr wants to merge 3 commits into
Conversation
…cket for AI fallback Answers the Discord question in #550 on how to auto-label AI-unhandled queries and escalate them to a human agent ticket. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Organization UI Review profile: ASSERTIVE Plan: Pro Run ID: 📒 Files selected for processing (1)
📝 WalkthroughWalkthroughThis change updates a single documentation file describing the "Open a ticket with a human agent" flow action. The "Last Updated" date was changed from October 2025 to July 2026. A new use-case section was added, describing how to escalate unanswered AI assistant queries to a human agent by detecting a fallback phrase, labeling the original message, and opening a ticket using the original question as the ticket body. Changes
Sequence Diagram(s)Not applicable — this change is a documentation-only update with no code execution flow to diagram. Compact Metadata Related Issues: None specified Suggested Labels: documentation Suggested Reviewers: None specified Poem 🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
|
🚀 Deployed on https://deploy-preview-645--glific-docs.netlify.app |
…tion Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Removing backticks around @results.<result name from step 1> left raw angle brackets, which MDX parsed as an invalid JSX tag and failed the build. Restore backticks around just this placeholder. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Summary
Fixes #550
Adds a "Use Case: Escalating unanswered AI queries to a human agent" section to the
Open a ticket with a human agentdoc, answering the Discord question referenced in the issue:Label the incoming messagein the AI fallback path so unhandled messages get labeled automatically (e.g.AI_unhandled).Open a ticket with a human agentright after labeling, so those queries route to a human agent.The walkthrough covers:
Split by a result in the flowon the assistant's response to branch on that fallback phrase (or theOtherbranch)AI_unhandledright there - since noWait for Responseruns between the contact's original question and this point, the label still lands on the original question, not the AI's replyOpen a ticket with a human agentnode immediately after, using the original question as the ticket bodyCross-links to the existing File Search / OpenAI Assistants doc, the Split by a Result in the Flow doc, and the Label the incoming message doc (recently expanded in #644) rather than duplicating their content.
Verification
No Glific feature returns a built-in "AI couldn't answer" signal, so the guidance relies on the org configuring a consistent fallback phrase in the assistant's Instructions and branching on it - this matches how orgs actually do this today per the linked Discord thread. The claim that the label still applies correctly a few nodes after
Wait for Response(as long as no otherWait for Responseintervenes) is grounded inglific/glific'sadd_input_labelsbehavior verified while writing #644 (it labelscontext.last_message, which is only overwritten when a new inbound message resumes the flow).Test plan
Summary by CodeRabbit