graduation-qa

Validate graduation requirement QA reports against canonical wiki evidence and triage records.

26|1|Updated Jan 8, 2025
One-click install
npx skills add https://github.com/allcll/allcll-backend --skill graduation-qa
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: graduation-qa
Source: https://github.com/allcll/allcll-backend/tree/main/docs/agent-harness/skills/graduation-qa
Command: npx skills add https://github.com/allcll/allcll-backend --skill graduation-qa

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps resolve graduation requirement QA reports by forcing decisions to come from the canonical wiki, not from guesses or code reverse-engineering. It prevents mixing unrelated reports together and keeps corrections tied to the correct root cause, evidence, and triage record.

Core Features & Use Cases

  • Report Triage: Groups multiple user reports into root-cause batches so one underlying issue is handled once.
  • Canonical Verification: Checks expected graduation behavior against the wiki, handbook references, and snapshot drift evidence before any code change.
  • Validation and Logging: Requires replay, full-snapshot diffing, and triage record updates to close the QA loop cleanly.
  • Use Case: A user reports that a specific graduation rule is being evaluated incorrectly; this Skill determines whether the issue is data, code, policy conflict, or an unverified contradiction and directs the next step.

Quick Start

Analyze this graduation QA report against the wiki truth source, identify the root-cause batch, and tell me whether it is a data issue, code issue, policy contradiction, or needs human confirmation.

Frequently Asked Questions about graduation-qa

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

FAQPage Schema
How do I validate graduation requirement QA reports against canonical wiki sources?

To validate graduation requirement QA reports, check expected graduation behavior against canonical wiki references and snapshot drift evidence before making any code changes. This ensures corrections are tied to verified root causes rather than guesses or reverse-engineered code logic.

What is the best way to triage multiple miscomputed graduation reports into root-cause batches?

Triage miscomputed graduation reports by grouping multiple user complaints into root-cause batches, ensuring one underlying issue is handled once. This prevents mixing unrelated reports and keeps corrections tied to the correct triage records.

Why does my graduation rule evaluation need replay validation and snapshot diff checks?

Graduation rule evaluation requires replay validation and full-snapshot diffing to close the QA loop cleanly. These checks verify expected behavior against wiki-backed evidence before any code fix or closure decision is approved.

How do I determine if a graduation QA issue is a data error, code bug, or policy contradiction?

Determine if a graduation QA issue is a data error, code bug, or policy contradiction by analyzing the report against the wiki truth source. This process identifies the root-cause batch and directs whether the issue needs a code fix or human confirmation.

Can I fix graduation requirement errors without checking the canonical wiki handbook first?

Fixing graduation requirement errors without checking the canonical wiki handbook first is not recommended. Decisions must come from canonical wiki sources to prevent unverified contradictions and ensure corrections match the correct root cause and evidence.

What should I do when graduation requirement validation shows a snapshot drift but no code changes?

When graduation requirement validation shows snapshot drift without code changes, update the triage records and perform replay validation. This confirms whether the drift indicates an actual data issue or an expected-behavior discrepancy requiring human confirmation.