GA dealbot dashboard landing page

[BigLep]

Done Criteria

We have a landing page with these components:

• Date picker
  • Absolute time (UTC)
    • Fine to simplify to days (no hours or minutes)
    • Use ISO8061 format
  • Relative time
    • Use ISO8601 form for specifying periods
• Dropdown to switch between Calibration/Mainnet
• Storage Provider table
  • Columns
    • Provider id/name
    • Whether approved
    • Data Storage Success Rate
    • Data Storage Sample Count
    • Data Retention Fault Rate
    • Data Retention Proof Count
    • Retrieval Success Rate
    • Retrieval Sample Count
  • Numbers will be grey/green/red based on certain condition
    • Grey: not enough samples
    • Green: meeting acceptance criteria
    • Red: not meeting acceptance criteria
  • Clicking on provider opens the to be created "Provider Detail Page" (GA dealbot provider detail page #177)
• Error Breakdown table
  • Columns
    • Error Type: substring of error message up until the first ":"
    • Check (one of "Data Storage" or "Retrieval")
    • Count
    • Providers
    • % of Total
  • Click on a row will list the set of providers that have seen that error and their relative count and percentage
  • Clicking on one of the providers will load the provider detail page and jump to the Error table filtered to that error

Other notes:

• URLs should be bookmarkable: if I change the data range, that should be reflected in the query string.
• We will link to any public docs we create about our policies, metric definitions, etc.

Why Important

This allow SP Approval List managers (i.e., FOC WG SP support personnel) to see how various SPs are doing and whether they should be approved or not.

User/Customer

1. SP Approval List managers
2. PDP SPs to see where they stand in relationship to others

Notes

1. There are notes and mockups in https://linear.app/filoz/document/public-definition-of-thresholds-for-dealbot-ccf29ae32583 , but that may not have the latest understanding. Check with @BigLep as he is trying to use this issue to document what engineers ned to implement.
   • https://uploads.linear.app/06cbbb88-947f-4459-9483-ed44072cd0a1/79f3bc9a-3df1-4adf-bb07-148cfd111156/c4554967-0d68-44a3-bb0b-6f85837c2f27

2. There are a various frontend and backend steps to accomplish this work scope. They can be added as subtasks.

[BigLep]

2026-01-30 discussion between @BigLep, @SgtPooki , and @silent-cipher:

• Add backlog item: Add plausible for usage stats on who and how the dealbot is being used
  • Add plausible tracking #180
• Rough estimation in ideal dev days areas of new work
  • Pulling in pdp fault data (either directly from browser from public subgraph or via dealbot API that is querying subgraph or chain): 1
  • Allowing arbitrary day ranges (accounting for percentile metrics). This is for the backend API work: 1
  • New UI layout: 6-8
    • Day picker widget: ~2
    • Provider table: ~2
    • Error table: ~2
    • Other stuff: including setting up tests, deeplinking
  • Provider detail page (GA dealbot provider detail page #177 )~3 days of assembling based on existing dealbot widgets and new polished landing page in this issue
• Rough plan of attack:
  • @silent-cipher will attack things from the frontend (i.e., this issue and GA dealbot provider detail page #177). He is doing the reads/presentation from the DB.
  • @SgtPooki will focus on making sure the checks are doing the right thing thing and run with the frequency / amdount we need. He is doing the writes into the DB.
  • @BigLep will focus on documentation items (specifying right behavior and making sure anyone can read and understand what is happening)