菜单
Run or configure post-game analysis on completed game memory dumps. Use when setting up the analysis pipeline, customizing the analysis prompt, reviewing learnings files, or running manual analysis. Triggers on "analyze", "post-game", "learnings", "score", "review game", "what went wrong".
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Create or modify a game brain (decision engine) that subclasses BaseBrain. Use when implementing per-tick decision logic, adding roles, tuning thresholds, or writing the decide() method for a new game plugin. Triggers on "brain", "decide", "role", "threshold", "heuristic", "decision logic".
Configure or run the self-improving code evolution pipeline. Use when setting up evolution for a new game, customizing editable files, tuning the evolution prompt, running manual evolution, or managing the weekly fork/reset lifecycle. Triggers on "evolution", "evolve", "self-improving", "str_replace", "fork", "reset", "promote", "weekly cycle".
Create or modify an LLM harness that subclasses BaseHarness. Use when implementing LLM context building, directive parsing, system prompts, or post-game analysis hooks for a game plugin. Triggers on "harness", "LLM context", "system prompt", "directive", "narrator", "advisor".
Create or modify the game policy (control loop) that subclasses BasePolicy. Use when implementing perceive() to parse game engine observations, execute() to convert commands to game actions, or wiring brain + harness together. Triggers on "policy", "perceive", "execute", "control loop", "game engine bridge".
Create or modify game event triggers that subclass BaseEventDetector. Use when implementing event detection for LLM consultation, adding new trigger types, tuning priorities, or configuring debounce. Triggers on "trigger", "event detection", "consult", "LLM scheduling", "when to call LLM".
Create or update wiki knowledge pages for a game plugin or the framework. Use when writing strategy docs, skill guides, mechanics references, or updating the transferable knowledge base. Triggers on "wiki", "strategy doc", "skill doc", "mechanics doc", "game knowledge", "write documentation".
| name | cyborg-post-game |
| description | Run or configure post-game analysis on completed game memory dumps. Use when setting up the analysis pipeline, customizing the analysis prompt, reviewing learnings files, or running manual analysis. Triggers on "analyze", "post-game", "learnings", "score", "review game", "what went wrong". |
After each game, an analysis LLM reviews the memory dump and produces structured learnings that feed back into the next game.
Game ends → memory dump (JSON) → analysis LLM → learnings.json
Next game loads learnings → enriched system prompt → better decisions
Every game produces the same JSON schema:
{
"score": 7,
"what_worked": ["gathered resources efficiently"],
"what_failed": ["didn't transition roles early enough"],
"actionable_rules": ["IF resources > 50 AND tick > 200 THEN switch role"],
"biggest_mistake": "stayed in gatherer mode until tick 400",
"recommended_next_game_strategy": "transition earlier"
}
In your game_config.py, set ANALYSIS_SYSTEM_PROMPT:
ANALYSIS_SYSTEM_PROMPT = """\
You are analyzing a completed {game_name} game.
Review the memory dump and assess:
1. Resource management efficiency
2. Role transition timing
3. Map exploration coverage
4. Response to opponent actions
Respond with JSON (no markdown fences):
{{
"score": 7,
"what_worked": ["..."],
"what_failed": ["..."],
"actionable_rules": ["IF x THEN y"],
"biggest_mistake": "...",
"recommended_next_game_strategy": "..."
}}
"""
Set up _run_post_game_background() in your harness:
def _run_post_game_background(self, dump, filepath, trigger):
from framework.base_analysis import run_post_game_analysis
run_post_game_analysis(
dump, self._analysis_provider,
memory_dump_path=filepath,
system_prompt=ANALYSIS_SYSTEM_PROMPT,
)
python -c "
from framework.base_analysis import run_post_game_analysis
from framework.providers import create_analysis_provider
import json
provider = create_analysis_provider('bedrock')
with open('runs/my_game/abc123_memory.json') as f:
dump = json.load(f)
run_post_game_analysis(dump, provider, memory_dump_path='runs/my_game/abc123_memory.json')
"
framework.base_evolution.accumulate_cross_game_memory() tracks patterns
across games in runs/cross_game_memory.json:
python scripts/record_score.py --record --steps 2500 --game my_game
python scripts/record_score.py --avg 5 --game my_game
python scripts/record_score.py --history --game my_game
| File | Purpose |
|---|---|
framework/base_analysis.py | Analysis runner, prompt builder, synthesis |
framework/score_tracker.py | Score history, rolling avg, escalation |
scripts/record_score.py | CLI for score tracking |