What to do
AI-generated actions across all four output types. Execute directly from here.
All improvements
8 pending5.5h/week lost to manual status reporting
Mariam hand-compiles status across 3 clients weekly. It ladders to no goal and produces nothing durable.
Auto-generate weekly status from Jira + GitLab; retire the macro.
Onboarding goal under-resourced
18% of capacity on a 35%-weight goal. FLOW-204 is the only active onboarding work and it's blocked.
Reallocate one engineer toward onboarding next sprint.
Hold the rename (!410) to protect !412
!410 collides with the critical-path security MR. Merging first forces a costly rebase.
Hold !410 until !412 merges, then rebase.
Make !407 a required check
The first automated e2e flow can retire 6h/release of manual QA once de-flaked.
Pair to add explicit waits, then gate merges on the e2e suite.
Omar turned a recurring fire into a closed incident
#88 diagnosis → !405 fix with 8 tests that reproduce the 503 storm. Leverage up 6 points.
Recognize in the team channel; document the backoff pattern for other tenants.
Ahmed's security MR is the week's best work
Closed a real validation gap on the top goal with 14 tests and observability.
Call out in 1:1; use as the reference example for reliability work.
API testing gap across 3 work units
Amer scored below his baseline on API-related work units three times this sprint — a recurring skill gap.
Enroll in an API testing workshop; pair on the e2e harness.
Async-decision discipline for the team
Two MRs landed without linked decisions; the weekly sync logs none. A lightweight RFC habit would help.
Adopt a 1-paragraph decision-record template on critical MRs.