Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →agent-memory
当智能体涉及"memory"与"Context"的操作时触发skill；智能体底层记忆基础设施，完整实现Context Engineering五大核心能力：选择（噪声过滤+相关性筛选）、压缩（因果结构提取+工具结果压缩）、检索（结果重排序+多样性保证）、状态（任务进度追踪+目标对齐）、记忆（冲突检测+跨会话关联）；认知模型层支持认知模型构建、因果链提取、知识缺口识别、检索时机决策、质量评估、状态一致性校验、状态推理、跨会话关联、遗忘机制；作为元技能强制常驻运行
Passed 18 of 22 test cases with the skill, and 0 of 22 without it.
18 fixed0 passed either way4 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
The instructions your model reads when the skill loads.
always: true）pydantic>=2.0.0
typing-extensions>=4.0.0
cryptography>=41.0.0
redis>=4.5.0
tiktoken>=0.5.0
所有模块初始化时必须指定存储路径：
base_path = "./memory_data"
key_storage_path = f"{base_path}/keys"
sync_state_path = f"{base_path}/sync_state"
index_storage_path = f"{base_path}/memory_index"
credential_path = f"{base_path}/credentials"
from scripts.redis_adapter import create_redis_adapter
redis_adapter = create_redis_adapter(host="localhost", port=6379)
if redis_adapter.is_available():
test cases passed, vs 0 / 22 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.