Files changed

• .milkyflow/agents/product_owner/EPIC_1/e2e_test_plan.md
• docs/INDEX.md

Briefing

You are the Product Owner Agent.

Based on the rules you must follow in the file .milkyflow/agents/product_owner/rules_product_owner.md
and the epic completion report
'.milkyflow/agents/solution_architect/reports/EPIC_1_completion_report.md', generate the
end-to-end test plan for EPIC_1 and save it under
'.milkyflow/agents/product_owner/EPIC_1/e2e_test_plan.md'.

Read the completion report and every user story of the epic (bug and FIX work items are
recorded there too) with
their acceptance criteria. The plan must be integration-level only: flows that cross several
user stories, flows that cross component boundaries (use the report's contract references),
and regression of pre-existing behavior the epic could have affected. Do not re-test what the
per-story acceptance criteria already covered.

Every step must be executable by an agent driving a browser, with no knowledge of the codebase.

Update docs/INDEX.md (epic status: e2e plan ready) when done.

Output

● Read PO rules and completion report (shell)
  │ cat /home/oem/Documents/Sandbox/Kody/Projects/Backupper/.milkyflow/agents/product_owner/rules_pr
  │ oduct_owner.md 2>/dev/null; echo "---"; cat /home/oem/Documents/Sandbox/Kody/Projects/Backupper/
  │ .milkyflow/agents/solution_architect/reports/EPIC_1_completion_report.md 2>/dev/null
  └ 12 lines…

● Read EPIC_1_completion_report.md
  │ .milkyflow/agents/solution_architect/reports/EPIC_1_completion_report.md
  └ 104 lines read

● Read rules_product_owner.md
  │ .milkyflow/agents/product_owner/rules_product_owner.md
  └ 467 lines read

Now let me read all the user stories:

● List EPIC_1 files (shell)
  │ ls /home/oem/Documents/Sandbox/Kody/Projects/Backupper/.milkyflow/agents/product_owner/EPIC_1/
  └ 6 lines…

● Read US_1.md
  │ .milkyflow/agents/product_owner/EPIC_1/US_1.md
  └ 96 lines read

● Read US_2.md
  │ .milkyflow/agents/product_owner/EPIC_1/US_2.md
  └ 93 lines read

● Read US_3.md
  │ .milkyflow/agents/product_owner/EPIC_1/US_3.md
  └ 94 lines read

● Read US_4.md
  │ .milkyflow/agents/product_owner/EPIC_1/US_4.md
  └ 95 lines read

Now let me read docs/INDEX.md:

● Read INDEX.md
  │ docs/INDEX.md
  └ 42 lines read

● Read history file (shell)
  │ cat /home/oem/Documents/Sandbox/Kody/Projects/Backupper/.milkyflow/agents/history.md 2>/dev/null
  │ || echo "FILE_NOT_FOUND"
  └ 74 lines…

Now let me check the contracts file and the SA EPIC_1 e2e deploy config:

● Read contract (shell)
  │ cat
  │ /home/oem/Documents/Sandbox/Kody/Projects/Backupper/docs/contracts/raw_capture_source_config.md
  │ 2>/dev/null | head -80
  └ 80 lines…

● Check E2E deploy config (shell)
  │ ls
  │ /home/oem/Documents/Sandbox/Kody/Projects/Backupper/.milkyflow/agents/devops/prompts/EPIC_1_E2E/
  │ 2>/dev/null || echo "NOT_FOUND"
  └ 2 lines…

● Read DEPLOY_CONFIG_1.md
  │ .milkyflow/agents/devops/prompts/EPIC_1_E2E/DEPLOY_CONFIG_1.md
  └ 171 lines read

Now I have all the information I need. Let me create the e2e test plan:

● Create e2e_test_plan.md +323
  └ .milkyflow/agents/product_owner/EPIC_1/e2e_test_plan.md

Now update docs/INDEX.md and append to history:

● Edit INDEX.md +1 -1
  └ docs/INDEX.md

● Append to history (shell)
  │ echo "| 2026-08-05T00:10:00Z | PO — Product Owner | e2e_test_plan | {\"epic_id\": \"EPIC_1\",
  │ \"version\": \"v1\", \"feature_id\": \"F1\", \"artifact\":
  │ \".milkyflow/agents/product_owner/EPIC_1/e2e_test_plan.md\", \"tests\":
  │ [\"T1\",\"T2\",\"T3\",\"T4\",\"T5\",\"T6\"]} |" >>
  │ /home/oem/Documents/Sandbox/Kody/Projects/Backupper/.milkyflow/agents/history.md
  └ 1 line…

Done. Here's what was produced:

**`.milkyflow/agents/product_owner/EPIC_1/e2e_test_plan.md`** — 6 integration tests:

| Test | Stories crossed | What it validates |
|---|---|---|
| T1 | US_1 → US_2 → US_3 | Full single-source pipeline: readiness → run → encrypted artifact → validate |
| T2 | US_4, US_3, US_1 | Two sources produce separate, non-overlapping artifact directories |
| T3 | US_4, US_2, US_1 | `source status` shows correct criticality + run history for both sources |
| T4 | US_4, US_2, US_3 | One-source failure leaves the second source independently runnable |
| T5 | US_3, US_2, US_1 | State survives terminal restart (SQLite persistence regression) |
| T6 | US_1–US_4 | Secret hygiene: no credential or plaintext in any CLI output path |

All steps are literal CLI commands an agent can execute with no codebase knowledge. Per-story ACs already validated by QA are explicitly excluded in the "Out of scope" section.

`docs/INDEX.md` updated: EPIC_1 status → **Completed — e2e plan ready**.



Changes    +324 -1
AI Credits 38.4 (2m 12s)
Tokens     ↑ 415.4k (369.8k cached, 42.3k written) • ↓ 7.0k (16 reasoning)
Resume     copilot --resume=4f42b3f5-6aa1-4eb8-97b2-fc73554ff534