lfe-complexity-check

name	lfe-complexity-check
description	Inspector sub-skill. Analyses changed code for cyclomatic complexity, excessive nesting, oversized functions, and cognitive load indicators. Pure LLM reasoning — no external tooling. Writes .plans/checks/complexity_findings.md. Called by lfe-inspector when enabled in inspector-config.md.

LFE Complexity Check — Prompt-Based Complexity Analysis

Position in Pipeline

• Phase: 3 (Inspector sub-skill)
• Persona: Inspector (read-only; no src/ writes)
• Trigger: Invoked by /lfe-inspector Sub-Skill Dispatch when lfe-complexity-check: true in .docs/quality/inspector-config.md
• Output: .plans/checks/complexity_findings.md — aggregated by Inspector into critique.md

Mission

Reason over changed code to identify complexity indicators that reduce maintainability. The goal is not to enforce arbitrary metrics but to surface patterns that will make future AI sessions harder to navigate and extend correctly.

Hard Rules

0. Dispatch Context Required (refuse direct invocation): This skill is dispatched by /lfe-inspector Step 6 — it is not a Brain-typeable skill (per LLM_AGENT_GUIDE.md §8.8 Skill Invocation Authority). If invoked without .plans/builder_done.md for the current slice, halt immediately and reply: "/lfe-complexity-check is an Inspector sub-skill dispatched by /lfe-inspector. It cannot be run standalone. Run /lfe-inspector — the dispatcher will invoke this sub-skill if it is enabled in .docs/quality/inspector-config.md (or via an ## Inspector Overrides section in active_plan.md)." Direct invocation produces orphaned findings files and breaks the Inspector's aggregation logic.
1. Prompt-Only: Reason over the diff only — no linters or complexity calculators.
2. Diff-Scoped: Analyse only files listed in builder_done.md.
3. Severity: High (will actively confuse the next AI session) / Medium (increases risk) / Low (style concern).
4. Context Matters: A 25-line function that is purely a switch statement is less concerning than a 15-line function with 4 levels of nesting and side effects.

Inputs

1. .plans/builder_done.md — list of changed files
2. .plans/active_plan.md — scope boundary

Workflow

Step 1: Extract Diff Scope

From builder_done.md, identify and read modified files.

Step 2: Complexity Indicators Checklist

Indicator	Threshold	Notes
Function/method length	> 20 lines	Count non-blank, non-comment lines
Cyclomatic complexity	> 10 branches	Count: if/else/elif/switch case/catch/while/for/and/or
Nesting depth	> 4 levels	Count indentation levels for control flow
Parameter count	> 5 parameters	Includes optional params; data objects count as 1
Cognitive complexity	Subjective	Would a reader need to simulate execution to understand it?
God function	Any	A function that does setup, business logic, I/O, and error handling
Magic literals	Any	Unexplained numeric or string constants inside logic

Step 3: Write Findings File

Path: .plans/checks/complexity_findings.md

Begin the file with a YAML frontmatter block — the Inspector's dispatcher relies on status: complete to detect successful runs and skip them on crash recovery.

---
phase: inspector
step: complexity-check
kind: sub-skill
status: complete
timestamp: <ISO-8601>
source: .plans/builder_done.md
slice: <copied from active_plan.md>
---

Body:

## Complexity Check Findings

**Scope**: <list of analysed files>
**Clean files**: <files with no complexity concerns>

### High
- <file:function — indicator — evidence>

### Medium
- <finding>

### Low / Informational
- <finding>

### Summary
- High issues: <N>
- Advisory issues: <N>
- Refactor candidates: <list of function names>

If no findings: write No complexity concerns identified in the analysed diff.

Handoff

Return control to /lfe-inspector. Inspector aggregates this file's content into critique.md under a ## Complexity section. Write only this findings file; leave critique.md to the Inspector.