Menu
Analyze system design — structure, coupling, code smells, and high-risk hotspots. Purely graph-based, no code reads.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Root cause analysis — trace execution path to a failure, narrow candidates, read minimal source only at suspected failure points.
Generate narrative-first, importance-weighted documentation for a repo, system, or subsystem with a selective reference layer. Use --full for deeper module/class/method coverage.
Route to the right Ix skill or command for your task
Change risk analysis — blast radius, affected systems, and what to test. Depth scales with risk level; low-risk targets stop early.
Deep dive into a symbol, feature, or bug. Graph-first, minimal code reads, early stopping when sufficient evidence found.
Generate a risk-ordered implementation plan for a set of targets. Assesses blast radius per target, finds data flows between them, and produces a safe change sequence.
| name | ix-architecture |
| description | Analyze system design — structure, coupling, code smells, and high-risk hotspots. Purely graph-based, no code reads. |
| argument-hint | ["optional scope — path","subsystem name","or empty for whole system"] |
Check command -v ix first. If unavailable, stop and say so — this skill requires a graph.
Answer: how healthy is this system's design, where are the weak boundaries, and what should be improved? Never reads source code — structural analysis only.
Run in parallel:
ix subsystems --format llm
ix subsystems --list --format llm
The --list output is the canonical region ID enumeration. All subsequent scoped ix subsystems calls must use only IDs (or name tokens) that appeared in this list. Never pass a raw region ID that was not returned by --list.
If $ARGUMENTS is provided, resolve it against the --list output first:
--list, proceed with that ID.--list, do NOT call ix subsystems <argument> directly — instead filter the repo-wide --format llm results by path prefix or name string.After resolving the target region, run:
ix subsystems <resolved-id> --explain
ix subsystems <resolved-id> --format llm
Unknown-target recovery: If a scoped ix subsystems call returns unknown_target or an ambiguous result, immediately fall back to filtering the full ix subsystems --list output. Do not retry with variant spellings or raw numeric IDs not in the enumeration.
Extract:
crosscut_score > 0.1 -> cross-cutting concern (design smell)confidence < 0.6 -> fuzzy boundary (uncertain region)ix smells --format llm
If $ARGUMENTS scopes to a path or subsystem, do not rerun ix smells with a scope flag. Filter the repo-wide results by path prefix after retrieval.
Classify each finding: orphan / god-module / weak-component.
Run only if Phase 1 or 2 revealed significant issues:
ix rank --by dependents --kind class --top 10 --exclude-path test --format llm
ix rank --by dependents --kind function --top 10 --exclude-path test --format llm
Correlate: are the most-depended-on entities also in poorly-bounded subsystems? These are the highest-risk components.
## Architecture Analysis
### System Structure
[Region hierarchy with file counts. Flag: low-confidence boundaries, high-coupling regions, cross-cutting modules.]
### Health Scores
| Region | Cohesion | Ext. Coupling | Boundary Ratio | Flag |
|--------|----------|---------------|----------------|------|
| [name] | [0-1] | [0-1] | [ratio] | [⚠ if bad] |
### Code Smells
**High severity:**
- [smell type] in [file/module] — [why it matters]
**Medium severity:**
- ...
### Hotspots
[Top components where structural debt + centrality combine — highest risk for changes]
### Improvement Areas
1. [specific issue] — [concrete suggestion]
2. ...
### What's healthy
[Briefly note well-structured areas — not everything is a problem]
Evidence: All claims must cite graph data (cohesion scores, smell counts, rank positions). No speculative design advice without structural evidence.