Decision Ladder

Package scope and prerequisites

This catalog ships instructions, not the executable runtime. Python modules, scripts, local services and optional tools named below must be installed separately from the upstream project. Check their availability before running commands; if they are missing, explain the prerequisite instead of claiming execution or results. The host agent may use model tokens while following these instructions.

Ponytail-inspired YAGNI enforcement for this toolkit. Before writing ANY code, climb this ladder. Each rung that passes saves the cost of every rung above it.

The Ladder

1. not_needed — Is the requirement explicitly unnecessary?
2. existing_module — Does the real codebase already solve it?
3. stdlib — Is it in Python's standard library?
4. native_platform — Does the browser, OS, database, or protocol provide it?
5. installed_dependency — Is an accepted dependency already available?
6. regex_oneliner — Does one readable expression suffice without cutting safety?
7. new_code — Only after affected files and the real flow were inspected.

Usage

from skills.decision_ladder.ladder import climb, audit_task_list

# Single task
result = climb("extract function names from a Python file")
# → LadderResult(rung="stdlib", solution="ast module (ast.parse/walk)", saved_lines=15)

# Audit a task list
report = audit_task_list([
    "parse JSON config",
    "design auth middleware",
    "count word frequency in text",
    "strip HTML tags from string",
    "implement custom OR-Tools solver",
])
# Inspect the report from the actual run; no avoided work or line count is established by this example.

Integration

Add to workflow-check as a pre-code hook. See hook.py.

Metrics

Tracks:

• % of tasks that could be avoided (stdlib, regex, existing module)
• Estimated lines saved per task
• Confidence score per suggestion