‹ 首页

agent-system-memory-proactivity-ops

@mahsumaktas · 收录于 1 周前 · 上游提交 4 个月前

Audit and improve AgentSystem memory retention, proactive behavior, learning quality, and agent consistency. Use when bots forget context, skip memory writes, wait for explicit prompts, or repeat the same errors across runs.

适合你,如果智能体常忘记上下文或需要反复提示。

/ 通过 npx 安装 校验哈希
npx oh-my-skill add mahsumaktas/agent-evolution-kit/agent-system-memory-proactivity-ops
/ 通过 bash 安装
curl -fsSL https://oh-my-skill.com/install.sh | bash -s -- mahsumaktas/agent-evolution-kit/agent-system-memory-proactivity-ops
/ 已经装过?验证本机副本,不用重装
npx oh-my-skill verify mahsumaktas/agent-evolution-kit/agent-system-memory-proactivity-ops
安装目标可用 --agent / --scope 或 --to 明确指定;省略时只会在唯一已存在的 agent 目录上自动选择,零命中或多命中会停止并提示。content_hash 缺失或不一致均拒装。
1GitHub stars
~263上下文体积 · 单文件
索引托管

怎么用

商店整理自技能原文 · 版本 c599c87 · 表述以原文为准
它做什么

装上后,Claude 会主动检查并修复记忆系统的健康度,确保它记住上下文、主动写入记忆、避免重复错误。

什么时候触发

当机器人忘记上下文、跳过记忆写入、等待显式提示或跨运行重复相同错误时触发。

装好后可以这样说
运行基线检查脚本。
执行 remediation 脚本。
验证修复效果。
技能原文 SKILL.md作者撰写 · MIT · c599c87

AgentSystem Memory Proactivity Ops

Use this skill to keep memory and proactivity loops healthy after config, prompt, or plugin changes.

Quick Start
  1. Run baseline checks:
bash scripts/proactivity_check.sh --hours 24
  1. Apply known remediation pack:
bash /Users/user/clawd/scripts/agent-system-plan2-remediate.sh
  1. Re-check and smoke-test memory jobs:
bash scripts/proactivity_check.sh --hours 6 --smoke
Workflow
  1. Verify scheduler contract and agent consistency metrics.
  2. Verify learning score, filler count, and output quality in recent memory jobs.
  3. Check latest runs of:
  4. hachi-self-compound (046bb2db-42ad-4b79-96d4-f9aa1392ede8)
  5. memory-fact-extraction (079a9ab8-d706-465c-801d-741f4a2d55a3)
  6. Validate learning artifacts were updated recently.
  7. If quality is below target, tighten prompt constraints and re-run smoke checks.
Resources
  • scripts/proactivity_check.sh: one-command audit + optional memory smoke tests.
  • references/quality-gates.md: acceptance thresholds and remediation sequence.
按 MIT 许可原样转载,未经改动 · 在 GitHub 查看 →

评论

登录即可评论;带「已验证安装」的,是发布者名下有本店的安装或持有记录。