brownfield-intakelisted

# Brownfield Intake VibeFlow's SDLC starts at REQUIREMENTS, which assumes a requirements doc to score. On a **greenfield** project you write a PRD. On a **brownfield** project you already have code, and you're almost never rebuilding the whole product — you're doing an **increment** (a feature, refactor, bug-fix, migration). This skill is the on-ramp: it makes VibeFlow *understand the existing code first*, then guides you to define the increment, and produces the scoped requirements artifact the rest of the SDLC runs against. ## Phase Contract Runs in **REQUIREMENTS** only (it produces the requirements artifact). If `currentPhase` is not REQUIREMENTS, emit a one-line note and stop. Writes are confined to `docs/**`, `.vibeflow/**`, and `vibeflow.config.json`. ## Step 1: Understand the codebase (so you don't have to explain it) Reuse `onboard`'s `.vibeflow/reports/codebase-fingerprint.md` if present; otherwise build one. Use the **codebase-intel MCP** (read-only) against the project root: - `mcp__codebase-intel__ci_analyze_structure` — languages, frameworks, test runners, build tools, module/package layout. - `mcp__codebase-intel__ci_dependency_graph` `{ root: ".", detectCycles: true }` — the import graph + cycles (so the increment respects real boundaries). - `mcp__codebase-intel__ci_find_hotspots` — churn × complexity hotspots. - `mcp__codebase-intel__ci_tech_debt_scan` — debt markers near the work. Write/refresh `.vibeflow/reports/codebase-fingerprint.md` (stack s