PR #14: add monitor-freshness workflow for ingest/reachability drift

[khalidsaidi]

Summary

• add a new scheduled workflow: .github/workflows/monitor-freshness.yml
• checks public https://ragmap-api.web.app/rag/stats (no secrets needed)
• fails if freshness drifts beyond thresholds:
  • lastSuccessfulIngestAt older than 36 hours
  • lastReachabilityRunAt older than 18 hours
• runs every 6 hours, offset by 30 minutes: 30 */6 * * *
• supports manual workflow_dispatch

Why

• prevent silent operational regressions from making freshness-based UX appear broken
• complements reachableMaxAgeHours and 4x/day reachability refresh cadence

Validation

• workflow YAML reviewed
• threshold logic and failure messages verified in script body

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: ee6a4e38f3

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Compare freshness age in seconds, not truncated hours

This workflow floors elapsed time to whole hours (/ 3600) before checking the >36 and >18 thresholds, so stale data can pass for up to 59 extra minutes. For example, an ingest timestamp 36h59m old still computes ingestAgeHours=36 and does not fail, which weakens the intended “older than 36h/18h” monitoring guarantees.

Useful? React with 👍 / 👎.