SuperAPI AI Directory

spec-test-plan

Testing & quality

View repository

Updated 10-16 01:48

About this skill

Spec Test Plan (V1: Test Plan)

Generate/update the test plan for the current Spec Pack: freeze the scope, strategy, environment, entry/exit criteria, risks, and verification checklist, and keep it traceably consistent with the acceptance criteria of requirements/* .


Inputs and Gates (must be followed)

REQUIRED SUB-SKILL: currently executing spec-context to obtain context, and echo in the conversation FEATURE_DIR=...(allows (reuse))。

  • required project-level memory:project/memory/product.md、project/memory/tech.md、project/memory/glossary.md; if it cannot be read, you must write CONTEXT GAP。
  • minimum requirement-level input:{FEATURE_DIR}/requirements/solution.md or {FEATURE_DIR}/requirements/prd.md at least one of them must exist; otherwise stop and write CONTEXT GAP。

Outputs (written to disk)

  • {FEATURE_DIR}/verification/test-plan.md

Mandatory Template (must be followed)

  • test-plan.md must be generated according to the template (do not invent structures/field names):
    • assets/test_plan_template.md

Content Requirements (minimum structure)

test-plan.md must contain at least:

  • Executive Summary (capability under test/version/goal/key risks)
  • Test Scope (In/Out)
  • Test Strategy (types: functional/UI/integration/regression/security…; methods: positive/negative/boundary)
  • Environment and Data (environment, accounts/permissions, data preparation method)
  • Entry Criteria
  • Exit Criteria (must include the criterion for "blocking delivery")
  • Risk and Verification Checklist (Owner/deadline/signal/action)
  • Traceability links (pointing to solution/prd 's AC/scope sources; if there is #impact-analysis , link it as well)
  • CONTEXT GAP (if any: impact caused by missing inputs)

DoD Self-Check (V1-DoD)

  • Scope In/Out is explicit and consistent with requirements/* consistent
  • has entry/exit criteria, and can clearly state "under what circumstances delivery is blocked"
  • Risks are not a mere checklist: each risk has a minimum verification action (Owner/signal/action)
  • has a traceability entry point: at least it can answer "which AC/scope definition is this plan based on"