Port TEA commands into workflows and preload Murat knowledge (#660)

* Port TEA commands into workflows and preload Murat knowledge

* Broke the giant knowledge dump into curated fragments under src/modules/bmm/testarch/knowledge/

* Broke the giant knowledge dump into curated fragments under src/modules/bmm/testarch/knowledge/

* updated the web bunles for tea, and spot updates for analyst and sm

* Replaced the old TEA brief with an indexed knowledge system: the agent now loads topic-specific
  docs from knowledge/ via tea-index.csv, workflows reference those fragments, and risk/level/
  priority guidance lives in the new fragment files

---------

Co-authored-by: Murat Ozcan <murat@mac.lan>
This commit is contained in:
Murat K Ozcan
2025-09-30 15:19:55 -05:00
committed by GitHub
parent 30fb0e67e1
commit df0c3e4bae
51 changed files with 1139 additions and 901 deletions

View File

@@ -0,0 +1,9 @@
# CI Pipeline and Burn-In Strategy
- Stage jobs: install/caching once, run `test-changed` for quick feedback, then shard full suites with `fail-fast: false` so evidence isnt lost.
- Re-run changed specs 510x (burn-in) before merging to flush flakes; fail the pipeline on the first inconsistent run.
- Upload artifacts on failure (videos, traces, HAR) and keep retry counts explicit—hidden retries hide instability.
- Use `wait-on` for app startup, enforce time budgets (<10 min per job), and document required secrets alongside workflows.
- Mirror CI scripts locally (`npm run test:ci`, `scripts/burn-in-changed.sh`) so devs reproduce pipeline behaviour exactly.
_Source: Murat CI/CD strategy blog, Playwright/Cypress workflow examples._