When does an AI agent use this pattern?

When the trigger described above matches the situation the agent is handling. The agent queries FusedFrames at runtime, this pattern is returned and the agent follows the recorded reasoning and steps to reach the recorded outcome.

How were the steps captured?

By the desktop app on the machine of someone who actually does this work. It records clicks, keystrokes, copy and paste, drag and drop, selected text and screen regions across Intercom, then groups them into the meaningful steps shown above.

Where does the reasoning come from?

From the operator. When the desktop app sees a decision the action alone cannot explain, it asks why in the moment. The operator's answer is recorded and becomes the reasoning attached to the pattern, so the AI agent gets the thinking and not just the steps.

What if my team handles this differently?

Capture how your team actually does it. The desktop app records what your operator does on your tools and the web app distills that into a pattern shaped by your team. Each team's library reflects their own approach rather than the example shown here.

What if a case looks like this pattern but does not quite fit?

The pattern is linked to 1 alternative. The agent can follow the connections shown on this page to a related pattern that matches the case more closely.

How does my AI agent actually call this pattern?

Through the read-only REST API or the npm CLI. Create an API key in the dashboard, choose whether it accesses all libraries or specific ones and have your agent query at runtime. The response is structured JSON ready for tool use.

Is the data shown on this page from a live customer?

No. Example libraries on this site are illustrative. Patterns in your own account are built from your team's captured actions and stay private to your workspace.

How do I build a pattern like this for my team?

Install the desktop app on the machine of the person who already does this work, ask them to handle a real case end to end, then distill the captured action into a pattern in the web app. The pattern you end up with mirrors how your team actually operates.

Resolving an Intercom conversation with a help article

When a customer's question matches a known help-centre article, the support rep sends the relevant snippet inline with the article URL using a saved-replies macro. The conversation is tagged 'self-served' for weekly reporting and closed five minutes later if no follow-up arrives.

Tags

intercomself-servehelp-centremacrosdeflection

Lets your agent filter to the right pattern

Categories and tags scope the retrieval set, so your agent receives only the patterns relevant to the task instead of scanning the full library.

What and why

The observed behaviour and the reasoning behind it.

Behaviour

When the customer's question matches a known help-centre article (password reset, invite a teammate, change billing email, export a report, configure SSO basics), the support rep opens the team's saved-replies macro in Intercom, picks the macro that maps to the article, sends it inline with the article URL, then waits five minutes before closing the conversation. The reply pastes the relevant snippet into the chat itself rather than only linking to the article, and the conversation is tagged 'self-served' so the deflection shows up in weekly reporting.

Reasoning

Most product questions in Intercom have already been answered in the help centre, and the customer just hasn't found the article. Pasting the relevant lines inline makes the answer readable on a phone where opening a link is friction. The five-minute wait before closing exists because customers often have a follow-up question seconds after reading the snippet, and closing prematurely sends a cold 'this conversation has ended' message which costs the support rep a CSAT point. The 'self-served' tag is what feeds the deflection rate metric the team reports on every Friday.

Gives your agent an operator's reasoning

Behaviour records the action; reasoning records the rationale. Your agent applies both to handle cases that diverge from the captured examples.

Cause and effect

What initiates this pattern and what it produces.

Trigger

An Intercom conversation tagged with a non-bug intent contains a question that maps cleanly to an existing help-centre article.

Outcome

The customer receives a complete answer inline in the conversation along with a link to the full article. The conversation is tagged 'self-served' and closed five minutes later if no follow-up arrives.

Tells your agent when to act and what to verify

Your agent stays within the scope set by the trigger and outcome, so it doesn't extend beyond the work or finish before the outcome is met.

Standard operating procedure

Step-by-step instructions to reproduce this pattern.

Intercom

Open the conversation from the inbox and confirm the intent tag is one of: billing, login, onboarding or account-change.

Bug-tagged conversations should never be deflected to a help article because the article will not fix anything. If the conversation looks like a bug but is tagged as something else, retag before deciding the response.

Expected: You are looking at a conversation whose intent could plausibly be answered by an existing help article.

Intercom

Open the saved-replies macro panel and search for the macro by topic keyword (e.g. 'invite', 'sso', 'export').

The team keeps macros named for the customer's likely keyword, not the internal feature name. 'invite' returns the macro for 'add a teammate' even though the feature is officially called 'organisation members'.

Expected: The relevant macro is selected in the panel and a preview of its body is visible.

Intercom

Insert the macro into the reply field and check that the help-centre URL points at the customer's locale.

Macros default to /en/ but the team's German and French customers prefer /de/ and /fr/. Swap the locale segment in the URL before sending if the customer's account is set to a non-English locale.

Expected: The reply contains the article snippet inline and a localised link to the full article.

Intercom

Send the reply and then add the 'self-served' tag to the conversation immediately.

Tagging on send rather than on close makes sure the conversation still gets counted as self-served even if the customer comes back with a follow-up that pulls the conversation into a longer thread.

Expected: The reply is delivered and the conversation carries the 'self-served' tag.

Intercom

Wait five minutes before closing. If the customer replies in that window, switch to a live response and stop the deflection flow.

Five minutes is the team-agreed window. Customers who come back within five minutes are usually telling you the article didn't actually answer their question, and continuing to insist on the article is the fastest way to a low CSAT score.

Expected: The conversation is closed cleanly five minutes after the deflection reply, or it is reopened and handled live.

Hands your agent your team's exact steps

Your agent executes the documented procedure in order, against the applications named at each step, producing consistent output across human and agent runs.

Book a demo

Related patterns

How this pattern connects to other patterns in the library.

Often previous

Triaging an Intercom conversation by topic and intent

Alternative to

Converting an Intercom conversation into a Zendesk ticket

Lets your agent confirm the match and its scope

Alternative patterns let your agent compare candidates and select the closest match. Previous and next patterns define position in the workflow, so the agent operates within the scope of this one.

Supporting actions

Actions that provide evidence for this pattern.

Sent SSO basics macro to Brightline Health, tagged self-served

Replied with invoice-export snippet to Periwinkle Retail and closed

Pasted teammate-invite article inline for Sundermark Capital

Sent password-reset macro to Koru Fitness in /de/ locale

Self-served Heron Insurance billing email change in chat

Shows your agent the evidence behind the pattern

Each pattern is derived from captured action sequences. Your agent can reference the source actions to verify any field, so nothing it acts on is authored by hand or hallucinated.

Metadata

Timestamps and identifiers.

Evidence	Observed 142 times across 5 connections
Applications	Intercom
First seen	4 Feb 2026, 09:08
Last seen	6 May 2026, 16:51

Helps your agent rank patterns by confidence

Action count and last-seen timestamp weight each pattern by frequency and recency, so your agent selects the highest-scoring match rather than the first returned.

Questions

Frequently asked questions

Book a demo View all examples

Speak to the founder

Get a demo. Watch a live capture, then an AI agent query the result.

Ask anything. Pricing, security or integrating with your stack.

No purchase obligation

Book a call

Start capturing

Record in minutes. Install once and work as normal.

Plug AI agents in. One API call from any AI agent stack.

Refund on unused credits if you cancel