Meta MuseUndisclosed (tool listing; no reported user results)Public · Free

A read-only 'find my money' connector for Muse: settlements, recalls, points

harris-ryderGitHub · 2026-09-19

Read original

Dowser is a read-only MCP connector for Muse that surfaces open class-action settlements, product recalls with refunds, unused card credits and more.

Business model

Dowser is an open-source read-only MCP connector for Muse that surfaces open class-action settlements, product recalls with refunds, unused card credits and expiring points, with a ready-made find-my-money prompt. It is a tool/adaptation listing: no user results have been published and AgentFlow has not tested it.

Dowser plugs into Muse as a read-only connector that matches you against open settlements, recalls, card perks and expiring points; claiming stays manual on official channels.

Workflow

  1. Connect Dowser to Muse as an MCP connector per the repo instructions. (AgentFlow reference flow)
  2. Run the find-my-money query to match open settlements, recalls and expiring points. (AgentFlow reference flow)
  3. Verify eligibility and claim each item on official channels only. (AgentFlow reference flow)

Adaptation plan

No user results exist for this tool; hit rate and payouts are unknown. The date shown is the aggregator's filing date. AgentFlow has not independently deployed this adaptation.

Starter prompt

Written by AgentFlow; not the original author’s prompt. Not independently deployed in Muse or Grok.

Run a find-my-money scan with Dowser: open class-action settlements, product-recall refunds, unused card benefits and expiring points matched to me. List each with amount, eligibility and the official claim link. Do not submit any claim for me - give me the list first.

Running costs and pitfalls

No user results exist for this tool; hit rate and payouts are unknown. The date shown is the aggregator's filing date. AgentFlow has not independently deployed this adaptation. No independent cost benchmark is available. Check subscriptions, permissions and external actions before running a workflow.

Author-reported results are not independently audited.

The breakdown and starter prompt are AgentFlow’s analysis and adaptation; results are not guaranteed.

Original text and media remain attributed to their authors.