14-day proof sprint checklist¶
Use this sprint to turn interest into visible proof, contributors, and stars.
Sprint goal¶
- Publish 5+ real proof entries in
docs/proof-log.md - Receive 3+ structured value-proof issues
- Improve star/contributor conversion with evidence-first updates
If you need a zero-friction start, use day1-proof-starter.md.
Daily loop (10–20 minutes)¶
For each day, complete:
- Run/collect one meaningful ship/no-ship example (from your work or a user report)
- Open/update one value-proof report issue
- Add/update one proof-log entry (
docs/proof-log.md) - Link any relevant PR/workflow run
- Close with one sentence: what changed because of this signal?
Day-by-day plan¶
Days 1–3 (bootstrap)¶
- Seed 2 proof-log entries from recent real runs
- Publish one issue using
value_proof_report.ymlas an example - Verify README links to proof flow are visible and working
Days 4–7 (consistency)¶
- Add at least 1 proof entry every day
- Convert at least 1 open issue/discussion into a proof report
- Tag recurring failing steps and note patterns
Days 8–11 (optimization)¶
- Identify the top 2 recurring friction points
- Update docs/quickstart wording to reduce one friction point
- Record before/after effect in proof log
- Run
make adaptive-premergeat least once and logjudgment_summary+ confidence
Days 12–14 (wrap + publish)¶
- Produce sprint summary (what improved, what still fails)
- Add “next sprint focus” bullets
- Post summary in GitHub Discussions or Releases notes
KPI tracker (fill daily)¶
| Date | New proof entries | New proof issues | Ship signals | No-ship signals | Stars (net) | Notes |
|---|---|---|---|---|---|---|
| YYYY-MM-DD | 0 | 0 | 0 | 0 | 0 | |
| YYYY-MM-DD | 0 | 0 | 0 | 0 | 0 | |
| YYYY-MM-DD | 0 | 0 | 0 | 0 | 0 |
Weekly review prompts¶
- Which failing step appears most often?
- Are users understanding remediation quickly?
- Which proof entry got the strongest response?
- What one README/docs change would improve conversion next week?
- Did adaptive reviewer confidence trend up or down this week?