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 Ramp, 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.

Rejecting a Ramp expense with an illegible or missing receipt

When a Ramp transaction arrives without a legible itemised receipt, the spend operations lead rejects it with reason 'missing_itemised_receipt' and asks the cardholder to retake the photo. A second offence in the same calendar month escalates the case to the senior RevOps manager for a coaching conversation.

Tags

spend-approvalramprejectionreceiptvat-recoveryescalation

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 a Ramp transaction arrives in the inbox without a legible itemised receipt, the spend operations lead rejects the transaction with the reason `missing_itemised_receipt`, sends a friendly Ramp comment asking the cardholder to retake the photo with line items visible and notes the rejection on the cardholder's spend record. A second rejection on the same cardholder in the same calendar month escalates the case to the senior RevOps manager.

Reasoning

VAT recovery requires the line items broken out, not just the total, for any spend above $50. A photo that shows the total alone passes the auto-receipt match but fails the VAT recovery and the team has to chase the receipt later under audit pressure, which is a worse experience than a polite rejection on the day. The second-offence escalation is a soft control: most cardholders fix the habit after one nudge and the small number who repeat get the senior manager's attention before the pattern becomes ingrained.

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

A Ramp transaction in the approval inbox has no receipt attached, has a receipt that does not show line items or has a receipt where the line items are illegible.

Outcome

The transaction is rejected with reason `missing_itemised_receipt`, the cardholder receives a Ramp comment explaining what to retake and the rejection is logged on the cardholder's spend record. A second offence in a calendar month surfaces to the senior RevOps manager for a quick coaching conversation.

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.

Ramp

Open the transaction and inspect the attached receipt.

Look for both totals and line items. Line items can be blurred or cropped out; totals can be present without line items. Either failure is a rejection. If the receipt is in a foreign language, translate the line items rather than rejecting; foreign receipts are fine as long as the line items are visible.

Expected: The receipt is fully reviewed for line items and totals.

Ramp

If the receipt is missing or illegible, click Reject and choose `missing_itemised_receipt` from the reason list.

Always pick the specific reason `missing_itemised_receipt`, never the generic `policy_violation`. The specific reason tells the cardholder what to fix; the generic reason makes them ask, which adds a back-and-forth nobody benefits from. Ramp also uses the specific reason to tune its receipt-OCR which improves over time.

Expected: The transaction status is Rejected with the specific reason.

Ramp

Add a friendly comment naming what is missing and what to retake.

The comment is the most important part of the rejection. Name the specific gap: 'Could you re-upload with the line items in shot? Totals look good but VAT recovery needs the breakdown.' Keep it under 30 words and never use the word `policy`. Frame it as helping with VAT recovery, not catching them out.

Expected: A short, friendly, specific comment is posted on the transaction.

Ramp

Log the rejection on the cardholder's spend record under the `repeat-tracker` tag.

The repeat tracker is how second offences surface. Without the tag a second offence in the same month looks like a new mistake; with the tag it shows up as a repeat for the senior RevOps manager to address. The tag clears at the start of each month so old offences do not haunt cardholders forever.

Expected: The rejection is tagged `repeat-tracker` on the cardholder's spend record.

Ramp

Check whether the cardholder has another rejection in the current calendar month.

Filter the cardholder's spend record by status Rejected and date this month. One rejection is normal even for careful cardholders; two in the same month is the threshold. Three is a coaching conversation; four is a card freeze. Most cardholders never see two.

Expected: Either no prior rejection exists this month or a second is identified.

Ramp

On a second offence in the same month, escalate to the senior RevOps manager via Ramp's escalation comment with the cardholder name and both transaction IDs.

Use Ramp's escalation comment, not Slack or email. The escalation comment carries the transaction history so the manager has everything in one place without switching apps. Include both transaction IDs and the cardholder name; the manager picks up the coaching conversation from there.

Expected: The senior RevOps manager has a Ramp escalation comment with both transaction IDs.

Ramp

Move to the next pending transaction.

Do not wait for the cardholder to reupload before continuing. Reuploads come back into the inbox as a fresh pending transaction in the same flow, so the inbox keeps moving. If the reupload is fine on second attempt the cardholder has had the experience of a quick correction rather than a slow review.

Expected: The inbox advances to the next transaction.

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

Approving a Ramp transaction with the right NetSuite GL code

Alternative to

Approving a Ramp transaction with the right NetSuite GL code

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.

Rejected Ramp txn_3RvZ1cKp ($142.18 Pret a Manger) reason missing_itemised_receipt

Posted retake-receipt comment on txn_3RvZ1cKp for cardholder Felix Nordstrom

Tagged 2nd-offence reject for cardholder Toby Marchant on repeat-tracker May

Escalated repeat-tracker case to Imogen Whitfield via Ramp escalation comment

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 22 times across 2 connections
Applications	Ramp
First seen	11 Feb 2026, 10:34
Last seen	5 May 2026, 11:48

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