resolving-scope-boundaries

name	resolving-scope-boundaries
description	Use when validating IDD scenarios - load .scope file, establish hard analysis boundary, index all in-scope source files

Resolving Scope Boundaries

Overview

Load the .scope file, read every listed source file, and establish a hard boundary for all subsequent analysis. Files not in scope do not exist for this validation run.

When to Use

• Second step in IDD validation pipeline
• After gherkin-parser has extracted scenarios
• Before code-mapper attempts to locate implementations

Core Process

Step 1: Load Scope File

Read every path listed in the === SCOPE FILES === section provided in the prompt.

For each path:

• If file: read directly, index metadata
• If directory: recursively find all .py, .ts, .tsx, .js, .jsx files (skip .spec.*, .test.*, .map)
• If unresolvable: log "UNRESOLVABLE: {path}", continue

Step 2: Index Each File

{
  "path": "relative/path/to/file.py",
  "language": "python | typescript | javascript",
  "line_count": 234,
  "functions": ["function_name"],
  "classes": ["ClassName"],
  "imports": ["module.path"],
  "status": "loaded | unresolvable"
}

Step 3: Declare Scope Boundary

Output this declaration before proceeding:

Scope boundary established.
Loaded: [N] files
  • path/to/file.py (234 lines)
  • ...

Unresolvable: [M] files
  • path/not/found.py

Hard Boundary Rules:
✓ Only files above are analyzed
✓ Imports to unresolvable files are not followed
✓ No prior-session knowledge used
✓ No codebase knowledge beyond loaded files

Hard Constraints

• Scope is immutable once declared — cannot be expanded mid-validation
• No transitive import following (imports pointing outside scope are noted but NOT followed)
• No cross-session knowledge from prior validation runs
• Scope declaration must appear BEFORE any code-mapper analysis begins

Output

Indexed scope boundary. All subsequent skills operate ONLY within this boundary.