Skip to content

Star Atlas Species Master Guide: Identification & Pipeline

This guide centralizes all identifiers and procedural logic for classifying the various species within the Star Atlas Galia Expanse.

1. Visual Identification Heuristics

MUD (Manus Ultima Divina) - Humans

  • Visuals: Standard biological humans.
  • Clothing: Tactical "PEARCE" gear, corporate uniforms, headsets.
  • Traits: High hair color variety (Red, Blue, Cyan, Multicolor).
  • Reference IDs: 001-002, 006-007, 013-026, 029, 031 (F, Red hair), 032 (F, Blue hair), 034 (F, Cyan hair, Monocle), 038 (F, Multicolor), 039 (M, Blonde), 040-051, 055, 060, 065, 070, 080, 090, 100, 110, 120, 130, 140, 150, 166, 169, 185, 195, 210, 235, 300.

Ustur (Sentient Androids)

  • Visuals: Robotic/mechanical humanoid constructs.
  • Key Features: Geometric plates, cyan/orange/blue core illumination. High armor variety with distinctive helmet types: Sleek, Bird-like/Avian (orange/yellow), Beetle-like, Insect-like, and Horned.
  • Reference IDs: 003, 004, 005, 008-011, 033, 035, 036, 167, 170-178, 180, 200, 225, 250, 290.

Sogmian (Warrior Philosophers)

  • Visuals: Noble alienoids, purple/pale blue skin.
  • Dimorfismo:
  • Sogmian Male: Smoother head, distinctive cranial structure.
  • Sogmian Female: Glowing red visors; head-tails/tentacles.
  • Reference IDs: 012 (F), 028 (F), 165 (M), 205. (Note: Initial scan of 160-200 range showed high density of Mierese, High Punaab and Ustur).

Mierese (Ethereal Oral storytellers)

  • Visuals: Lavender/lilac skin, slender frames, pointed ears.
  • Dimorfismo:
  • Mierese Male: Extremely long dreadlock-like head tentacles.
  • Mierese Female: Shorter braid-like tentacles, elaborate cranial crest.
  • Reference IDs: 160 (M), 163 (F), 164 (F), 168 (M), 190 (F), 215 (F), 220 (F), 240 (F), 260 (F), 270 (F).

Punaab (Sentient Mammals) ⚠️ SCALE SENSITIVE

  • High Punaab: White/light tiger-patterned fur. Ornate gold mantles. NO tails. Scale: 60cm (Knee height).
  • Profound Punaab: Grey/dark fur. Small furry constructs. NO tails. Tactical/industrial gear with monocles. Scale: 40cm (Calf height).
  • Reference IDs:
  • High: 161, 162, 198, 280.
  • Profound: 027, 030, 037.

Tufa (Metagenic Swarm)

  • Visuals: Biomechanical and mineral hybrids; crystalline shell structures.
  • Key Features: Strange geometric forms, pulsing energy cores.

Photoli (Beings of Light)

  • Visuals: Extra-galactic lifeforms made of pure essence/energy.
  • Schism: 7 Dark Photoli (arrogant predatory exiles who consume essence). Mainstream Photoli = benevolent teachers with hidden agenda to access Iris.

To manage the acervo of 350+ unique crew members, assets are organized through a dedicated script-driven pipeline.

Current Inventory Status (Feb 2026)

  • Total identified assets: 350
  • Total classified assets: 101 (28.8% progress)
  • Classification Pace: Lotes de 10-20 imagens via inspeção visual.
  • Range Observations: Lower ranges (000-100) are predominantly MUD (humans). Range 160-200 shows high density of Ustur, Mierese, and Punaab.
  • Mierese: 10
  • Ustur: 25
  • Sogmian: 4
  • Punaab (High + Profound): 7
  • MUD: 55

⚠️ Goal: Complete classification for the remaining 249 assets to reach 100% coverage.

Detailed ID Lookup (Species by ID Range/List)

  • Ustur: [3, 4, 5, 8-11, 33, 35, 36, 167, 170-178, 180, 200, 225, 250, 290]
  • Sogmian: [12, 28, 165, 205]
  • Mierese: [160, 163, 164, 168, 190, 215, 220, 240, 260, 270]
  • High Punaab: [161, 162, 198, 280]
  • Profound Punaab: [27, 30, 37]
  • MUD: [1, 2, 6, 7, 13-26, 29, 31, 32, 34, 38-51, 55, 60, 65, 70, 80, 90, 100, 110, 120, 130, 140, 150, 166, 169, 185, 195, 210, 235, 300]

Directory Structure

star_atlas_crew/
├── all_crews/             # Raw harvested JPEG assets
│   ├── crew_000XXX.jpeg   # Padded naming convention (001-300)
│   └── crew_X.jpeg        # Non-padded naming convention (1-50)
├── species/               # Classified assets
│   ├── mud/ | ustur/ | sogmian/ | mierese/ | high_punaab/ | profound_punaab/
├── species_classifier.py  # Primary sorting script
├── classifications.json   # Unified metadata mapping (Species/Gender)
└── scene_generator.py     # Production pipeline for character scenes

Classification Workflow

  1. Audit: Visual identification using the heuristics in Section 1.
  2. Hardcoded Mapping: Update KNOWN_CLASSIFICATIONS dictionary in species_classifier.py.
  3. Execution: Run species_classifier.py to physically move files and generate classifications.json.
  4. Scene Generation: scene_generator.py queries the JSON to select characters and inject specific "Grimdark" descriptors.