🛡️ ShipGuard Pro — The Quality Gate for AI-Built Software
Don’t trust “Done.” Verify it. AI coding agents write code fast, but fast doesn’t mean ready. ShipGuard Pro is an evidence-first AI verification skill designed to audit AI output before it turns into a production issue.

🔍 What It Does
Acting as a skeptical principal engineer, QA lead, and security reviewer, ShipGuard asks:

Was the requirement satisfied?

Is there real evidence the feature works?

Are tests evaluating actual behavior?

Did implementation drift, or is the AI claiming false completion?

⚙️ Verification Modes

Full Audit & Task Verification: Complete health and requirement checks.

Pre-Ship & PR Gates: Decision engine (🟢 SHIP, 🟡 WARNINGS, 🔴 DO NOT SHIP, ⚫ INSUFFICIENT EVIDENCE) and change impact analysis.

Claim Verification: Validates claims like “fixed,” “secure,” or “tested.”

📊 Grades Evidence (E0–E5)
ShipGuard relies on strict evidence levels—from E0 (AI assertion) to E5 (Requirement + implementation + runtime + edge cases). Missing evidence is never treated as a pass.

🛠️ Scope & Capability
Evaluates functional correctness, test depth, security, reliability, performance, and requirement drift.

🤖 Built for AI Workflows
A portable SKILL.md agent skill compatible with Claude Code, Cursor, Codex, Gemini CLI, and more.

📦 Portable & 🚫 No API keys required

🧾 Machine-readable evidence ledgers & Markdown/HTML reports