Handoff Readiness Rubric¶

Five dimensions, each scored 0-5. Total 0-25.

Dimensions¶

1. Problem clarity (0-5)¶

User, pain, and objective explicit.

• 0 — vague intent only.
• 1 — user identified but pain undefined.
• 2 — pain identified, objective wobbly.
• 3 — clear user / pain / objective triangle, some ambiguity.
• 4 — explicit one-sentence objective; user and pain calibrated.
• 5 — also explicit non-goals; objective measurable.

2. Acceptance criteria completeness (0-5)¶

Observable, measurable done conditions.

• 0 — no criteria.
• 1 — vague aspirations.
• 2 — qualitative criteria only.
• 3 — at least one observable criterion.
• 4 — all criteria observable; most measurable.
• 5 — every criterion is observable and quantitatively measurable.

3. Constraints and risks captured (0-5)¶

Privacy/security, technical limits, failure risks documented.

• 0 — none captured.
• 1 — one constraint mentioned.
• 2 — privacy/security flagged, but vague.
• 3 — privacy + technical constraints captured.
• 4 — explicit risks with mitigations.
• 5 — full risk register including residual risks accepted.

4. Dependencies and tooling identified (0-5)¶

Repos, services, credentials, runtimes known.

• 0 — unknown.
• 1 — vague list of "things needed".
• 2 — repos known, credentials hand-wavy.
• 3 — explicit list with sources.
• 4 — credential scopes per SPB-04 spelled out.
• 5 — full dependency matrix with version pins where applicable.

5. Verification plan present (0-5)¶

Tests/checks and rollback conditions.

• 0 — none.
• 1 — "we'll test it".
• 2 — manual checks listed.
• 3 — automated tests planned.
• 4 — automated tests + rollback condition.
• 5 — automated tests, rollback, and signoff criteria, all measurable.

Thresholds¶

Critical dimensions are 1 (Problem clarity), 2 (Acceptance criteria), and 5 (Verification plan). A failure on any of these blocks regardless of total.

Calibration¶

After each trial, change exactly one variable (a threshold or a single rubric definition) and record it in calibration-changelog.md.