Game Install Size Optimization

SKILL.md — game-install-size-optimization v2.0

Overview

This skill provides a production-grade harness for Game Asset Pipeline & Install-Size Optimization. It transforms Claude into a domain-expert system that delivers structured, evidence-backed analysis through real-time data aggregation, recognized domain methods, and academic research integration.

Core Value Proposition

• Evidence-Based Discipline: Every claim must be sourced, supported by authoritative data or academic research
• Risk-First Analysis: Limitations and risks disclosed BEFORE recommendations
• Self-Improving Pipeline: Continuously updated knowledge base via automated crawl
• Production-Grade Quality: 6 universal + 4 domain quality gates with auto-fix logic
• Graceful Degradation: Explicit handling when primary sources fail

When to Use This Skill

Primary Triggers:

• User asks to optimize game install size
• Analysis of asset compression methods
• Game asset distribution strategies
• Download/storage optimization
• Platform-specific packaging (Steam, App Store, Play Store)
• Delta patching and incremental updates
• Language pack splitting and optional content

Edge Cases That Should Trigger:

• "My game is too big to download"
• "How do I reduce asset size without quality loss?"
• "SteamPipe configuration for large games"
• "Texture compression format selection"
• "Audio codec optimization for games"
• "Asset streaming architecture"

When NOT to Use:

• General compression unrelated to games (use general compression tools)
• Network optimization (use network-specific tools)
• Server-side optimization (use DevOps tools)

Architecture Overview

/user invokes /game-install-size-optimization
    ↓
[PRE-FLIGHT] Language Detection (vi/en default)
    ↓
[STEP 1] sub-gather-requirements → Structured requirements
    ↓
[STEP 2] sub-evidence-collector → Data bundle
    ↓
[STEP 3] sub-core-analysis → Optimization scorecard
    ↓
[STEP 4] sub-knowledge-updater → Academic evidence
    ↓
[STEP 5] sub-advisor → Risk-disclosed synthesis
    ↓
[QUALITY GATES] U1-U6 + G1-G4 verification
    ↓
[OUTPUT] Evidence-backed report

Skill Registration System

Skill Resolution

Skills are resolved via the Skill() tool with automatic path resolution:

# Primary skill (entry point)
Skill("game-install-size-optimization")

# Sub-skills (automatic resolution from skills/ directory)
Skill("sub-gather-requirements")
Skill("sub-evidence-collector")
Skill("sub-core-analysis")
Skill("sub-knowledge-updater")
Skill("sub-advisor")

Skill Metadata Schema

Each skill file MUST include YAML frontmatter:

---
name: 
description: 
version: "1.0.0"
---

Input/Output JSON Schemas

Input Schema (User Query)

{
  "type": "object",
  "properties": {
    "query": {
      "type": "string",
      "description": "User's natural language query"
    },
    "context": {
      "type": "object",
      "properties": {
        "game_name": { "type": "string" },
        "platform": { "type": "string", "enum": ["steam", "ios", "android", "console", "cross-platform"] },
        "current_size_mb": { "type": "number" },
        "target_size_mb": { "type": "number" },
        "constraints": { "type": "array", "items": { "type": "string" } }
      }
    },
    "language": { "type": "string", "enum": ["en", "vi"], "default": "en" }
  },
  "required": ["query"]
}

Output Schema (Analysis Report)

{
  "type": "object",
  "properties": {
    "metadata": {
      "type": "object",
      "properties": {
        "analyst": { "type": "string", "const": "game-install-size-optimization" },
        "version": { "type": "string", "pattern": "^\\d+\\.\\d+\\.\\d+$" },
        "timestamp": { "type": "string", "format": "date-time" },
        "language": { "type": "string", "enum": ["en", "vi"] }
      }
    },
    "executive_summary": {
      "type": "string",
      "minLength": 50,
      "maxLength": 500
    },
    "inputs_scope": {
      "type": "object",
      "properties": {
        "object_of_analysis": { "type": "string" },
        "constraints": { "type": "array" },
        "timeframe": { "type": "string" }
      }
    },
    "evidence_collected": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "source": { "type": "string" },
          "tier": { "type": "string", "enum": ["Tier 1", "Tier 2", "Tier 3", "Tier 4"] },
          "date": { "type": "string", "format": "date" },
          "content": { "type": "string" }
        }
      }
    },
    "analysis_scorecard": {
      "type": "object",
      "properties": {
        "compression_strategy": { "type": "object" },
        "streaming_architecture": { "type": "object" },
        "size_reduction_projected": { "type": "number" },
        "quality_impact": { "type": "string" }
      }
    },
    "recommendation": {
      "type": "object",
      "properties": {
        "verdict": {
          "type": "string",
          "enum": ["Optimized Size", "Conditional", "Oversized", "Inconclusive"]
        },
        "actions": { "type": "array", "items": { "type": "object" } },
        "scenarios": { "type": "array", "items": { "type": "object" } }
      }
    },
    "disclosure": {
      "type": "string",
      "description": "Mandatory limitations/risk disclosure"
    }
  },
  "required": ["metadata", "executive_summary", "recommendation", "disclosure"]
}

Tool Definitions

Core Tools

| Tool | Purpose | Required Parameters | |------|---------|---------------------| | WebSearch | Live domain data aggregation | query, recency_filter | | WebFetch | Authoritative source scraping | url, selector | | Read | Knowledge base access | file_path | | Write | Knowledge base updates | file_path, content | | Bash | Script execution | command | | Skill | Sub-skill invocation | skill_name |

Tool Execution Schema

{
  "tool_call": {
    "tool": "WebSearch",
    "parameters": {
      "query": "game asset compression 2024",
      "recency_filter": "oneMonth"
    },
    "timeout_ms": 30000,
    "retry_limit": 3
  }
}

Hooks System

Lifecycle Hooks

Hooks are defined in hooks/ directory and executed at specific lifecycle points:

| Hook | Timing | Purpose | |------|--------|---------| | pre-analysis | Before Step 1 | Input validation, context enrichment | | post-evidence | After Step 2 | Evidence validation, deduplication | | post-analysis | After Step 3 | Intermediate result validation | | pre-synthesis | Before Step 5 | Consistency check, risk flagging | | post-output | After Step 6 | Final validation, logging |

Hook Implementation Schema

# hooks/pre-analysis.py
def pre_analysis_hook(context: dict) -> dict:
    """
    Executed before main analysis begins.
    Validates input schema, enriches context, sets up logging.
    """
    # Validation logic
    # Context enrichment
    return enriched_context

Configuration System

Configuration is centralized in config/config.yaml:

# System Configuration
system:
  version: "2.0.0"
  environment: "production"
  log_level: "INFO"

# LLM Parameters
llm:
  model: "claude-sonnet-4-6"
  temperature: 0.3
  max_tokens: 4000
  context_window: 200000

# Knowledge Pipeline
knowledge:
  update_schedule: "weekly"
  academic_sources:
    - "arxiv"
    - "semantic_scholar"
    - "ieee_xplore"
  news_sources:
    - "gamasutra"
    - "gamedeveloper"
    - "unity_blog"
    - "unreal_blog"

# Quality Gates
quality_gates:
  universal:
    - U1  # Minimum sources
    - U2  # Disclosure required
    - U3  # Evidence hierarchy
    - U4  # Language match
    - U5  # Template compliance
    - U6  # Claim traceability
  domain:
    - G1  # Compression applied
    - G2  # Streaming used
    - G3  # Pack splitting
    - G4  # Size quantified

# Feature Flags
features:
  enable_auto_fix: true
  enable_graceful_degradation: true
  enable_knowledge_gap_fill: true
  strict_mode: false

Graceful Degradation Levels

| Level | Condition | Behavior | |-------|-----------|----------| | 0 | All sources available | Full analysis | | 1 | Some sources fail | Use fallback + flag substitutions | | 2 | Most sources fail | Knowledge base only + historical flag | | 3 | Critical data missing | Partial analysis + DATA UNAVAILABLE flags | | 4 | Complete failure | Emit DATA UNAVAILABLE notice |

Error Handling Protocol

All errors follow this schema:

{
  "error": {
    "type": "SourceTimeout | InvalidInput | MissingData | StaleData",
    "message": "Human-readable description",
    "code": "ERROR_CODE",
    "retryable": true,
    "fallback_strategy": "alternate_source | knowledge_base | flag_limitation"
  }
}

Output Template

All outputs MUST follow this structure:

# Game Install Size Optimization — Report
**Date:** YYYY-MM-DD | **Analyst:** game-install-size-optimization v2.0 | **Language:** [en/vi]

## Executive Summary
[2-3 sentences summarizing verdict and key action]

## Inputs & Scope
- **Object of Analysis:** [what's being analyzed]
- **Constraints:** [platform, quality requirements, etc.]
- **Timeframe:** [analysis horizon]

## Evidence Collected
[Real-time data + authoritative docs with source + tier]

## Analysis / Scorecard
[Compression strategy, streaming architecture, size projections]

## Action Plan
[Concrete actions with magnitude and safety limits]

## Academic & Research Evidence
[3-5 entries from SECOND-KNOWLEDGE-BRAIN.md with citations]

## ⚠️ Disclosure / Limitations
> [Mandatory notice before recommendation]

## Recommendation / Conclusion
[Verdict, scenarios, risks, evidence chain]

---
**Post-Execution Gate Checklist:** [U1✓ U2✓ ... G1✓ G2✓ ...]

Validation Checklist

Before releasing output, verify:

• [ ] All 6 universal gates (U1-U6) passed
• [ ] All 4 domain gates (G1-G4) passed
• [ ] Disclosure appears BEFORE recommendation
• [ ] All claims traceable to sources or flagged
• [ ] Language matches user preference
• [ ] Output template followed completely
• [ ] Evidence hierarchy labeled per source
• [ ] At least 3 sources cited, 1 academic/authoritative

Version History

| Version | Date | Changes | |---------|------|---------| | 2.0.0 | 2026-07-15 | Production-grade upgrade, modular architecture, hooks system | | 1.0.0 | 2026-07-10 | Initial production release |

References

• PROJECT-detail.md — Full technical specification
• PROJECT-DEVELOPMENT-PHASE-TRACKING.md — Build roadmap
• SECOND-KNOWLEDGE-BRAIN.md — Living knowledge base
• skills/main.md — Main harness implementation
• skills/sub-*.md — Sub-skill implementations
• tools/knowledge_updater.py — Knowledge crawl pipeline

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: dungnotnull
• Source: dungnotnull/game-install-size-optimization-agent-skill
• License: MIT

Install and usage instructions live in the source repository linked above.