| name | report |
| description | Build the monthly Thunderbird support report. Usage: /report <month> <year> (e.g. /report may 2026) |
Monthly Support Report — $month $year
Run the full report cycle for $month $year. Follow every step in order.
Step 1 — Re-fetch all live data sources
Do NOT use YAML values from a prior session. Always fetch fresh.
FeatureOS ideas (board 17437):
featureos-cli posts list --query "sort=votes_count&order=desc&per_page=25&status=all" --json
Update both new_this_month and top_alltime in data/$month_$year.yaml.
new_this_month: cross-reference with Lisa's FeatureOS board filtered by creation date for this monthtop_alltime: top 5 by votes_count from CLI output- JSON key is
feature_requests, not posts
- Ideas can be removed from the board between sessions — verify all entries still exist
SUMO data (Roland's reports):
- Desktop:
https://thunderbird.github.io/thunderbird-metrics-and-reports/html_reports/desktop/$year-$month_num-sumo-desktop-report.html
- Android:
https://thunderbird.github.io/thunderbird-metrics-and-reports/html_reports/android/$year-$month_num-sumo-android-report.html
- Fetch both pages and update
sumo: section in the YAML
Step 2 — Confirm Play Store CSVs are in place
Required files in project root or data/input/:
reviews_reviews_net.thunderbird.android_$year$month_num.csvreviews_reviews_com.fsck.k9_$year$month_num.csv- Optional beta:
reviews_reviews_net.thunderbird.android.beta_$year$month_num.csv
CSVs are UTF-16 encoded. Use all rows — no date filtering. The export is scoped by last activity date, not submit date.
If a file has a (1) suffix (macOS duplicate), update csv_tb_stable in the YAML.
Step 3 — Confirm Zendesk metrics are in the YAML
These come from Lisa's MoM spreadsheet. If overall_csat is null, ask Lisa before proceeding.
Check for:
zendesk.overall_csat, donor_csat, tbpro_csatzendesk.total_tickets, donor_tickets, tbpro_ticketszendesk.replies_to_low_star (Zendesk is authoritative — covers all 3 apps)zendesk.tbpro_csat_note if TB Pro sample is low
Step 4 — Run the generator
uv run scripts/generate.py $month $year
This auto-populates:
- Friction point quotes, devices, languages (from CSVs)
- K-9 Discourse forum data (fetched live)
- 3-month trends for friction points and SUMO metrics
- History updated in
data/history.json
Step 5 — Fill in remaining narrative sections
These cannot be auto-populated — fill in lisa/$year/$month.md:
- Lede — 2-3 sentences. What's the headline story? CSAT trend, volume surge, K-9 signal, etc.
- Community Support key signals —
[Key signal from Roland's data] and [Key signal] placeholders
- Meeting 4★+ Goal bullets — 3 priorities with stats, trends, and why they matter
- Receive / Resolve / Resound — what came in, what worked, what to celebrate
- Experiments & Iterations — Lisa fills in manually
Check the K-9 Discourse data and Play Store K-9 analysis for signals worth calling out in the lede or K-9 Churn Watch section.
Step 6 — Verify dashboard
Open lisa/$year/$month.html and check:
- All stat cards show values (no
— where data should be)
- TB Pro note renders as small italic, not a heading
- Quarter label on rating trend is correct (Q1=Jan-Mar, Q2=Apr-Jun, etc.)
- K-9 Forum tab shows topics, resolved %, unanswered %, top themes with 3-month trends and NEW ↑ badges, top contributors
- Theme count that more than doubles vs. the start of the 3-month window renders in red
- SUMO tables show MoM deltas and 3-month trends
Step 7 — Commit and push
Stage only report files — never the raw CSV exports:
GIT_COMMITTER_EMAIL="lisajill@users.noreply.github.com" GIT_COMMITTER_NAME="Moment" \
git commit --author="Moment <lisajill@users.noreply.github.com>" \
-m "Add $month $year support report" \
data/$month_$year.yaml lisa/$year/$month.md lisa/$year/$month.html \
lisa/$year/$month.csv data/history.json index.md
git push
Verify live at: https://thunderbird.github.io/thunderbird-support-reports/lisa/$year/$month.html
Key links
