DCS-Afterburner

A mission linting, diagnostics, and safe optimization toolkit for DCS World .miz files.

Browser-based mission analyzer → — drop a .miz (and optional dcs.log) directly in your browser. No install required, nothing uploaded.

Analyze missions before deployment, catch common performance killers, compare versions over time, and optionally apply low-risk automatic optimizations. Optionally load a dcs.log file alongside any mission to correlate static findings with real runtime errors.

---

What it does

DCS .miz files are ZIP archives containing Lua-based mission configuration. Afterburner unpacks them, inspects the contents, and runs heuristic checks against known performance and stability patterns.

---

Use cases

• Pre-release validation — scan a mission before putting it on the server
• Live issue diagnosis — load a dcs.log to find what broke during the last session
• Legacy cleanup — find obvious performance problems in old missions
• Before/after comparison — check if a new version got heavier or cleaner
• Safe optimization — apply low-risk cleanup automatically with a backup
• CI gate — fail a build if a mission exceeds thresholds or contains banned patterns

---

Installation

pip install dcs-afterburner

Or from source:

git clone https://github.com/your-org/dcs-afterburner
cd dcs-afterburner
pip install -e .

---

Usage

# Analyze a mission and print a summary
afterburner analyze mission.miz

# Output machine-readable JSON (for CI/CD or dashboards)
afterburner analyze mission.miz --json

# Generate a markdown report
afterburner report mission.miz --format md

# Apply safe optimizations (always creates a backup first)
afterburner optimize mission.miz --safe

# Compare two versions of a mission
afterburner diff old.miz new.miz

# Analyze with a DCS log file for runtime correlation
afterburner analyze mission.miz --log dcs.log

# Inspect a log file on its own
afterburner logs dcs.log

# List all available rules
afterburner rules list

# Explain a specific rule
afterburner rules explain PERF_001

---

What gets checked

Mission size and bloat

• Excessive unit, static object, trigger, and zone counts
• Oversized embedded script blocks
• Large archive payload and duplicate assets

Trigger system

• Excessive continuous / polling triggers (TIME MORE patterns)
• Duplicate trigger actions
• Missing or non-descriptive trigger names

AI and unit performance

• High density of active AI groups at mission start
• Large ground formations in small areas
• Route complexity above threshold
• Excessive late-activation groups

Scripts and Lua

• Oversized embedded scripts
• Duplicate script frameworks (MOOSE, MIST, CTLD bundled multiple times)
• Timer-heavy and loop-heavy code patterns
• Heavy event handlers without filtering

DCS log analysis (optional — supply a dcs.log)

• Lua errors and stack traces from the last session
• Scheduler spam and high-frequency timer abuse
• Repeated errors indicating runaway loops
• Framework load failures and nil dereferences
• Runtime confirmation of static findings

Multiplayer and server health

• Too many active slots relative to mission complexity
• Excessive radio menu items
• High object density near spawn areas

Maintainability

• Unnamed or badly named groups
• Duplicate group names
• Missing mission metadata

---

Configuration

Create an afterburner.yaml in your project to tune thresholds:

rules:
  max_active_ground_units: 250
  max_trigger_count: 150
  max_script_size_kb: 1024
  warn_on_unnamed_groups: true

optimize:
  safe_mode_only: true
  create_backup: true

output:
  format: markdown

---

Risk scoring

Each mission receives a score based on weighted findings:

Score	Band
92–100	LOW
75–91	MODERATE
50–74	HIGH
< 50	CRITICAL

Scores are always accompanied by an explanation. A score without reasons is not output.

---

Safe optimization

The --safe flag applies only low-risk, reversible transformations:

• Remove exact-duplicate embedded assets
• Strip orphaned or temporary files from the archive
• Optionally rename unnamed groups and triggers using a safe prefix scheme

The following are never modified automatically:

• AI routes
• Trigger logic
• Group or unit deletion
• Mission balance or tasking behavior

Every optimization run produces a change log listing each change as applied, skipped, or unsafe.

---

GitHub Actions

- name: Lint mission files
  run: |
    pip install dcs-afterburner
    afterburner analyze *.miz --json > report.json
    afterburner analyze *.miz --fail-on critical

---

License

MIT