mb-verify

Verify task completion against acceptance criteria and generate verification reports.

53|11|Updated Mar 5, 2026
One-click install
npx skills add https://github.com/mrvladd-d/memobank --skill mb-verify
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mb-verify
Source: https://github.com/mrvladd-d/memobank/tree/main/skills/mb-verify
Command: npx skills add https://github.com/mrvladd-d/memobank --skill mb-verify

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Verifying a completed task against acceptance criteria can be error-prone and non-repeatable. mb-verify provides a structured loop to prime the context, run deterministic checks, capture evidence, and produce a final VERDICT.

Core Features & Use Cases

  • Verification loop: read plan/context, compare AC/REQ, run checks and record evidence.
  • Evidence management: store logs, screenshots, and reproduction steps.
  • Output generation: create verification.md and update task backlog/backlog statuses on PASS/FAIL; plus create bug docs or follow-ups on failure.
  • Use Case: QA teams can rapidly confirm task readiness before release.

Quick Start

Run mb-verify TASK-123 to validate a completed task and generate its verification report.

Frequently Asked Questions about mb-verify

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I verify task completion against acceptance criteria?

To verify task completion against acceptance criteria, you can run a structured verification loop that compares task outputs to requirements, runs deterministic checks, and records evidence to produce an auditable PASS or FAIL verdict.

What is auditable task verification and how does it work?

Auditable task validation is a process that enforces alignment to plan and acceptance criteria sources by running deterministic checks, capturing reproducible evidence like logs and steps, and generating a final traceable PASS or FAIL verdict.

How do I generate reproducible evidence for QA task sign-off?

You generate reproducible evidence for QA task sign-off by executing a verification loop that captures logs, screenshots, and reproduction steps into a verification.md file, syncing task backlog statuses based on the final outcome.

Can I automate the creation of bug reports when a task verification fails?

Yes, when a task verification results in a FAIL verdict, the process automatically generates bug documents and follow-up tasks alongside the verification report to ensure failures are traceable and actionable.

Does the verification process require any external dependencies?

No, the verification process operates as a standalone loop without requiring external dependencies or components, directly reading plan context to compare acceptance criteria and update task backlog statuses.