# Architecture Drift Detection

> Compare current source code against the stored evidence-backed reverse-engineering baseline to identify stale architectural knowledge, undocumented routing changes, new integrations, removed behavior, or prior baseline errors.

- Skill: `tuana-vn/architecture-drift-detection` (Agent Skill)
- Install (CLI): `npx skillmds@latest add tuana-vn/architecture-drift-detection`
- Raw SKILL.md: https://api.skillmd.com/api/skills/tuana-vn/architecture-drift-detection/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: tuana-vn (https://skillmd.com/u/tuana-vn)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/tuana-vn/architecture-drift-detection

---


# Architecture Drift Detection

## Principle

The source is the primary truth.

The stored reverse-engineering baseline may become stale.

## Compare

Check high-value baseline claims against current source:

- inbound interfaces
- client implementations
- selectors/factories
- configuration
- target/model/mode branches
- integrations
- protocol boundaries
- retry/fallback
- error mappings

## Drift Matrix

| Baseline Claim | Current Source | Status | Severity | Evidence | Baseline Update Needed |
|---|---|---|---|---|---|

Status:

- NO DRIFT
- EXPECTED DRIFT
- UNDOCUMENTED DRIFT
- BASELINE ERROR
- UNKNOWN

## Output

`docs/reverse-engineering/drift/<baseline-or-date>-drift-review.md`

Do not automatically rewrite baseline. Report drift first for review.