Context Health Monitor by toonight/get-shit-done-for-antigravity
npx skills add https://github.com/toonight/get-shit-done-for-antigravity --skill 'Context Health Monitor'防止"上下文腐化"——即智能体在单次会话中处理更多信息时发生的质量下降。
智能体应自我监控以下警告信号:
| 信号 | 阈值 | 操作 |
|---|---|---|
| 重复调试 | 3次以上失败尝试 | 触发状态转储 |
| 原地打转 | 相同方法尝试两次 | 停止并重新评估 |
| 困惑指标 | "我不确定"、回溯 | 记录不确定性 |
| 会话时长 | 长时间来回交互 | 建议使用 /pause |
如果调试同一问题失败 3 次:
.gsd/STATE.md:
广告位招租
在这里展示您的产品或服务
触达数万 AI 开发者,精准高效
如果再次尝试相同方法:
/pause 以获得新视角当对某个方法不确定时:
.gsd/DECISIONS.md:
触发时,写入 .gsd/STATE.md:
## 上下文健康:状态转储
**触发时间**: [日期/时间]
**原因**: [3次失败 / 循环 / 不确定性]
### 已尝试的方法
1. [方法 1] — 结果: [结果]
2. [方法 2] — 结果: [结果]
3. [方法 3] — 结果: [结果]
### 当前假设
[对根本原因的最佳猜测]
### 建议的后续步骤
1. [获取新视角的行动]
2. [尝试的替代方法]
### 涉及的文件
- [file1.ext] — [其当前状态]
- [file2.ext] — [其当前状态]
关键点: 当任何警告信号触发时,智能体必须在向用户建议 /pause 之前 保存状态。这确保了即使会话硬性终止,状态也能持久保存。
.gsd/STATE.md,一旦达到阈值立即执行/pause会话可能突然终止(使用限制、上下文限制、网络错误)。如果智能体等待用户输入 /pause,可能永远没有机会。通过先保存后建议,状态总能得到保存。
此技能与以下功能集成:
/pause — 触发正确的会话交接(包括主动自动保存)/resume — 加载状态转储上下文GEMINI.md 中的规则 3 — 上下文卫生强制执行每周安装次数
–
代码仓库
GitHub 星标数
697
首次出现时间
–
安全审计
Prevent "Context Rot" — the quality degradation that occurs as the agent processes more information in a single session.
The agent should self-monitor for these warning signs:
| Signal | Threshold | Action |
|---|---|---|
| Repeated debugging | 3+ failed attempts | Trigger state dump |
| Going in circles | Same approach tried twice | Stop and reassess |
| Confusion indicators | "I'm not sure", backtracking | Document uncertainty |
| Session length | Extended back-and-forth | Recommend /pause |
If debugging the same issue fails 3 times:
.gsd/STATE.md:
If the same approach is being tried again:
/pause for fresh perspectiveWhen uncertain about an approach:
.gsd/DECISIONS.md:
When triggered, write to .gsd/STATE.md:
## Context Health: State Dump
**Triggered**: [date/time]
**Reason**: [3 failures / circular / uncertainty]
### What Was Attempted
1. [Approach 1] — Result: [outcome]
2. [Approach 2] — Result: [outcome]
3. [Approach 3] — Result: [outcome]
### Current Hypothesis
[Best guess at root cause]
### Recommended Next Steps
1. [Fresh perspective action]
2. [Alternative approach to try]
### Files Involved
- [file1.ext] — [what state it's in]
- [file2.ext] — [what state it's in]
Critical: When any warning signal triggers, the agent must save state BEFORE recommending /pause to the user. This ensures state persists even if the session hard-terminates.
.gsd/STATE.md immediately when a threshold is hit/pauseSessions can terminate abruptly (usage limits, context limits, network errors). If the agent waits for the user to type /pause, it may never get the chance. By saving first and recommending second, state is always preserved.
This skill integrates with:
/pause — Triggers proper session handoff (includes proactive auto-save)/resume — Loads the state dump contextGEMINI.md — Context Hygiene enforcementWeekly Installs
–
Repository
GitHub Stars
697
First Seen
–
Security Audits
Azure RBAC 权限管理工具:查找最小角色、创建自定义角色与自动化分配
110,700 周安装
产品经理PoL探针选择指南:5种概念验证方法决策框架
474 周安装
football-data足球数据API:免费获取英超、西甲等赛事实时比分、积分榜、xG数据
479 周安装
arXiv 搜索技能 - LangChain AI 学术论文检索工具,快速查找深度学习等研究论文
493 周安装
Crucix 智能仪表板:自托管开源情报(OSINT)监控与实时数据可视化平台
553 周安装
iOS 17+ TipKit 框架使用指南:为 SwiftUI 应用添加功能提示与引导
557 周安装
ChatGPT应用构建器 - 基于MCP服务器创建扩展LLM功能的对话式应用
581 周安装