Daily Amazon negative review triage with seller reply drafts

By General Input

Every morning, scan your Amazon product reviews, flag new one and two star ones, draft polite seller replies, and route the worst ones to your team.

Integrations

  • Real-Time Amazon Data
  • Linear
  • Slack

Type

Agentic Task

Categories

  • Customer Support
  • Operations

Build me an agent workflow that triages new Amazon negative reviews every morning. It should run on a cron at 8am in my local time zone, Monday through Friday.

Inputs the user configures once: a list of Amazon ASINs to monitor, the Amazon marketplace country (default US), the Linear team that should receive issues, the Slack channel for the digest, and a short paragraph of brand voice guidelines for the drafted replies.

For each ASIN, call Real-Time Amazon Data Get Top Product Reviews. This returns the eight publicly visible top reviews without needing a logged-in Amazon cookie, which is what makes it safe for an unattended cron. Keep a small piece of state (a list of review IDs already seen per ASIN) so the agent only acts on reviews that are new since the last run.

For each new one or two star review, the agent should: (1) classify the root issue into one of product quality, shipping or packaging, sizing or fit, unclear instructions, customer expectation mismatch, or defective unit; (2) draft a polite, specific seller reply in our brand voice that does not violate Amazon's review response policies (no asking for the review to be removed, no offering compensation in public, no personal contact info in the public reply, just empathy plus a clear next step); (3) decide a severity of low, medium, or high based on whether the complaint reads like a one-off, a known issue, or a recurring product defect.

For every medium or high severity review, call Linear Create Issue in the configured team. The issue title should be the ASIN plus a short summary of the complaint. The description should include the full review text, reviewer name and star rating, the chosen issue category, the severity, the drafted seller reply ready to copy into Seller Central, and a link to the product page on Amazon. Set Linear priority to High for high severity and Normal for medium.

At the end of the run, call Slack Send a Message to post a single digest into the configured channel. The digest should include: total reviews triaged, breakdown by issue category, breakdown by severity, and a bulleted list of the newly created Linear issues with their issue keys and titles linked. If there were no new negative reviews, post a short all-clear message instead.

Keep the agent focused and quiet. Do not open Linear issues for low severity reviews. Do not post per-review Slack messages, only the daily digest. Do not attempt to publish the reply to Amazon, it is meant to be human-reviewed.

Related prompts

Explore more prompts
Call overdue Xero customers with an AI collections agentWin back LiveChat visitors whose chats went unansweredChat quality review board for LiveChat support leadsWin back no-show and cancelled appointments every morningLive Loop returns analytics with product-level drill-downNewsletter pre-flight and approval board for Mailjet sendsTurn a prospect spreadsheet into personalized sequence enrollmentsMailjet email delivery lookup console for support teamsCatch feature flags that never got switched on in productionKajabi customer support console for member access fixes