Load when working with project_knowledge.json, context files, or managing knowledge cache. Provides patterns for efficient project context management with 90% token reduction.
📖 中文导读:Load when working with project_knowledge.json, context files, or managing knowledge cache. Provides patterns for efficient project context management with 90% token reduction. 以下正文为该技能的原始 SKILL.md 内容(英文),上方安装命令可直接使用。
name: knowledge
description: Load when working with project_knowledge.json, context files, or managing knowledge cache. Provides patterns for efficient project context management with 90% token reduction.
# Pattern 1: Session start - load knowledge
head -100 project_knowledge.json
# Pattern 2: Query specific entity (rare, only if cache miss)
python .github/scripts/knowledge.py --query "entity_name"
# Pattern 3: Update knowledge after session
python .github/scripts/knowledge.py --update
# Pattern 4: Using knowledge programmatically
import json
with open('project_knowledge.json') as f:
for i, line in enumerate(f):
if i >= 100:
break
data = json.loads(line)
if data.get('type') == 'hot_cache':
entity_refs = data['entity_refs']
elif data.get('type') == 'gotchas':
issues = data['issues']