首页
continual-learning
microsoft/skills
通过使用钩子、两级内存（全局和局部）以及自动模式检测，为人工智能编码代理实现了一个持续学习循环，以便在不同会话之间保留并应用所学知识。
...展开全部
40
更新时间 2026-09-10
人工智能编程代理的持续学习
您的智能体在不同会话之间会忘记所有内容。持续学习可以解决这个问题。
The Loop
Experience → Capture → Reflect → Persist → Apply
↑ │
└───────────────────────────────────────┘
快速入门
安装挂钩（一步完成）：
cp -r hooks/continual-learning .github/hooks/
首次会话时自动初始化。无需配置。
两级内存
全局（~/.copilot/learnings.db）——在所有项目中随您而动：
- 工具模式（哪些工具失效，哪些有效）
- 跨项目约定
- 通用编码偏好
本地（.copilot-memory/learnings.db）——仅适用于当前代码库：
- 项目特定约定
- 此代码库中的常见错误
- 团队规范
学习成果的存储方式
自动（通过钩子）
该钩子会监听工具的执行结果并检测失败模式：
Session 1: bash tool fails 4 times → learning stored: "bash frequently fails"
Session 2: hook surfaces that learning at start → agent adjusts approach
代理原生（通过 store_memory / SQL）
代理可直接写入学习结果：
INSERT INTO learnings (scope, category, content, source)
VALUES ('local', 'convention', 'This project uses Result not exceptions', 'user_correction');
分类： pattern, mistake, preference, tool_insight
手动（内存文件）
用于可读且受版本控制的知识：
# .copilot-memory/conventions.md
- Use DefaultAzureCredential for all Azure auth
- Parameter is semantic_configuration_name=, not semantic_configuration=
压缩
经验值会随时间衰减：
- 访问量较低且超过 60 天的条目将被清理
- 高价值的学习内容（被频繁引用）将永久保留
- 工具日志将在7天后被清理
这既能防止数据无限制增长，又能保留关键内容。
最佳实践
- 只需一步即可安装——如果耗时超过
cp -r，就难以被采纳 - 正确界定范围——工具模式采用全局范围，项目规范采用本地范围
- 具体明确——
"Use semantic_configuration_name="胜过"use the right parameter" - 让效果累积——每次工作时段的小改进，数周后将产生指数级的收益
在 GitHub 上查看
---
name: continual-learning
description: Implements a continual learning loop for AI coding agents using hooks, two-tier memory (global and local), and automatic pattern detection to persist and apply learnings across sessions.
---
# Continual Learning for AI Coding Agents
Your agent forgets everything between sessions. Continual learning fixes that.
## The Loop
```
Experience → Capture → Reflect → Persist → Apply
↑ │
└───────────────────────────────────────┘
```
## Quick Start
Install the hook (one step):
```bash
cp -r hooks/continual-learning .github/hooks/
```
Auto-initializes on first session. No config needed.
## Two-Tier Memory
**Global** (`~/.copilot/learnings.db`) — follows you across all projects:
- Tool patterns (which tools fail, which work)
- Cross-project conventions
- General coding preferences
**Local** (`.copilot-memory/learnings.db`) — stays with this repo:
- Project-specific conventions
- Common mistakes for this codebase
- Team preferences
## How Learnings Get Stored
### Automatic (via hooks)
The hook observes tool outcomes and detects failure patterns:
```
Session 1: bash tool fails 4 times → learning stored: "bash frequently fails"
Session 2: hook surfaces that learning at start → agent adjusts approach
```
### Agent-native (via store_memory / SQL)
The agent can write learnings directly:
```sql
INSERT INTO learnings (scope, category, content, source)
VALUES ('local', 'convention', 'This project uses Result<T> not exceptions', 'user_correction');
```
Categories: `pattern`, `mistake`, `preference`, `tool_insight`
### Manual (memory files)
For human-readable, version-controlled knowledge:
```markdown
# .copilot-memory/conventions.md
- Use DefaultAzureCredential for all Azure auth
- Parameter is semantic_configuration_name=, not semantic_configuration=
```
## Compaction
Learnings decay over time:
- Entries older than 60 days with low hit count are pruned
- High-value learnings (frequently referenced) persist indefinitely
- Tool logs are pruned after 7 days
This prevents unbounded growth while preserving what matters.
## Best Practices
1. **One step to install** — if it takes more than `cp -r`, it won't get adopted
2. **Scope correctly** — global for tool patterns, local for project conventions
3. **Be specific** — `"Use semantic_configuration_name="` beats `"use the right parameter"`
4. **Let it compound** — small improvements per session create exponential gains over weeks
所有文件
1 个文件
相关技能
algorithmic-art
更新时间 2026-08-27
systematic-debugging
更新时间 2026-09-03
tech-debt-tracker
更新时间 2026-08-29
receiving-code-review
更新时间 2026-09-03
