statico

name	statico
description	Rust CLI static analyzer for TypeScript and Rust projects. Detects dead code, unused exports/types/dependencies, circular dependencies, code duplication, and framework-specific issues. Supports AI-optimized output formats, auto-fix suggestions, and Mermaid dependency graphs. Use when analyzing code health, measuring refactoring impact, setting up CI code-quality gates, or exploring unfamiliar codebases.
trigger	statico, static analysis, dead code, unused exports, code health, refactoring impact, SARIF, code duplication, TypeScript analysis, Rust analysis, ai format, mermaid, pr-comment

statico — Static Analyzer for TypeScript & Rust

A Rust CLI tool that statically analyzes TypeScript and Rust projects for dead code, unused exports, circular dependencies, duplication, and more. Auto-detects frameworks (Next.js, Payload CMS, Angular, NestJS) for accurate entry-point-aware analysis.

Building

cd /Users/dom/programming/statico && cargo build --release --features deep-resolution
# Binary: target/release/statico

Commands

# Analyze a project (default: JSON to stdout)
statico analyze <path>

# Output formats
statico analyze <path> --format html        # Interactive HTML report (open in browser)
statico analyze <path> --format markdown    # Markdown summary for terminals/PRs
statico analyze <path> --format sarif       # SARIF for GitHub Code Scanning
statico analyze <path> --format ai          # Compact LLM-optimized JSON (~500 tokens)
statico analyze <path> --format context     # Ultra-compact one-liner (~100 tokens) for system prompts
statico analyze <path> --format mermaid     # Dependency graph as Mermaid diagram
statico analyze <path> --format pr-comment  # GitHub-ready PR comment with emoji tables

# Auto-fix suggestions (dry run)
statico analyze <path> --fix --dry-run      # Unified diff patches, safe to pipe to git apply --check

# Filters & CI
statico analyze <path> --min-confidence 0.8  # Suppress low-confidence findings
statico analyze <path> --exit-code            # Exit 1 if issues found (CI gating)

# Compare two snapshots (before/after refactoring)
statico diff before.json after.json

Language Support

Language	Dead Code	Unused Exports	Circular Deps	Duplication	Framework Detection
TypeScript	✅	✅	✅	✅	Next.js, Payload CMS, Angular, NestJS
Rust	✅	✅ (pub items)	✅	✅	—

When to Use

Situation	Action
Before/after refactoring	analyze → save JSON → refactor → analyze → diff
Code review	analyze --format markdown to spot new dead code or unused exports
PR comment	analyze --format pr-comment for GitHub-ready output
Adding features	Run analyze to verify no regressions introduced
CI/CD setup	--format sarif for GitHub Code Scanning, --exit-code for pass/fail
Exploring a codebase	--format html for an interactive overview
AI agent context	--format ai for LLM-optimized JSON, --format context for system prompts
Dependency visualization	--format mermaid for import graph with dead code islands
Auto-fix cleanup	--fix --dry-run to generate safe patches

Understanding Results

• Confidence scores: 0.95+ = high (unreachable from any entry point). ~0.7 = medium (not reachable from framework entry points but reachable from scripts/migrations/tests). Filter with --min-confidence.
• Issues are informational — not pass/fail. A codebase with 0 issues is unusual.
• Duplication % is informational. 10–20% is typical for mature codebases.
• Health score: 0–100, weighted by issue severity.

AI Output Formats

--format ai — LLM-Optimized JSON

Compact JSON (~500 tokens) with only actionable data. Issues ranked by impact (LOC wasted). Includes suggested action per issue.

{
  "summary": {
    "health_score": 53.9,
    "total_files": 412,
    "issue_counts": { "dead_code": 12, "unused_exports": 847 }
  },
  "top_issues": [
    {
      "type": "dead_code",
      "path": "src/legacy/utils.ts",
      "lines_of_code": 340,
      "confidence": 0.98,
      "impact": "high",
      "suggested_action": "Delete file — unreachable from any entry point"
    }
  ],
  "files_at_risk": [
    { "path": "src/legacy/index.ts", "issue_count": 14 }
  ]
}

--format context — Ultra-Compact Summary

~100 tokens. Designed for AGENTS.md injection or system prompt context.

Health: 53.9/100 | 412 files | Dead code: 12 (3,401 LOC) | Unused exports: 847 | Circles: 3 | Dup: 18.2% | Top risk: src/legacy/index.ts (14 issues)

--format mermaid — Dependency Graph

Mermaid diagram showing module dependencies, circular deps highlighted in red, dead code islands in gray.

--format pr-comment — GitHub PR Comment

GitHub-flavored Markdown with emoji tables, before/after comparison support, and actionable items grouped by severity.

--fix --dry-run — Auto-Fix Patches

Outputs unified diff patches for safe auto-fixes:

• Dead file removal patches
• Unused export removal hints
• Safe to pipe to git apply --check for validation

JSON Output Schema (default --format json)

{
  "version": "0.2.0",
  "$schema": "...",
  "summary": {
    "total_files": 0,
    "total_lines": 0,
    "total_exports": 0,
    "total_types": 0,
    "issue_counts": { /* per-category counts */ },
    "health_score": 0,        // 0–100
    "duplication_percentage": 0.0
  },
  "structure": { /* file tree, module graph */ },
  "dependencies": { /* import graph, dependency tree */ },
  "quality": { /* metrics per file */ },
  "issues": {
    "dead_code": [],
    "unused_exports": [],
    "unused_types": [],
    "duplicate_code": [],
    "gotchas": [],
    "circular_dependencies": [],
    "unused_dependencies": [],
    "duplicate_exports": [],
    "unresolved_imports": [],
    "unlisted_dependencies": []
  },
  "duplication": { /* clone pairs, percentages */ },
  "detected_frameworks": []  // e.g. ["nextjs", "payload-cms"]
}

Framework Detection

Statico auto-detects these frameworks and adjusts entry-point conventions for more accurate dead code analysis:

• Next.js — pages router, app router, API routes, middleware
• Payload CMS — config, collections, blocks, field-level hooks
• Angular — modules, components, services, guards, resolvers
• NestJS — modules, controllers, providers, guards, interceptors

AI Workflow Examples

One-command code health check (for AI agents)

/Users/dom/programming/statico/target/release/statico analyze . --format ai --quiet

Check if refactoring introduced dead code

/Users/dom/programming/statico/target/release/statico analyze ./src > /tmp/before.json
# ... make changes ...
/Users/dom/programming/statico/target/release/statico analyze ./src > /tmp/after.json
/Users/dom/programming/statico/target/release/statico diff /tmp/before.json /tmp/after.json

Generate context for system prompt injection

/Users/dom/programming/statico/target/release/statico analyze . --format context --quiet
# Paste output into AGENTS.md or CLAUDE.md

Auto-fix suggestions (review before applying)

/Users/dom/programming/statico/target/release/statico analyze . --fix --dry-run --quiet
# Review patches, then: git apply < patch.diff

PR comment for code review

/Users/dom/programming/statico/target/release/statico analyze . --format pr-comment --quiet
# Paste into GitHub PR comment

Dependency graph visualization

/Users/dom/programming/statico/target/release/statico analyze . --format mermaid --quiet
# Paste into any Markdown renderer that supports Mermaid

Generate a shareable code health report

/Users/dom/programming/statico/target/release/statico analyze ./src --format html --output report.html
# Share report.html with the team

Set up GitHub Actions CI

- name: Static analysis
  run: target/release/statico analyze ./src --format sarif --output results.sarif --exit-code
- name: Upload SARIF
  uses: github/codeql-action/upload-sarif@v3
  with:
    sarif_file: results.sarif