自我迭代内核：用复盘与分层记忆把 Agent 失误写成规则
自我迭代内核（@user_c0f5ae43/self-iteration-core）是 SkillHub 社区 ai-agent 类 Skill，v1.0.0，无需 API Key。它不是外部 API 工具，而是通过 SKILL.md 训练 Agent：翻车或被纠正时走复盘四步法（挖机制根因、单点改 MEMORY/TOOLS/SOP、edit_file 落地、验证后沉淀），并配合 MEMORY.md 与 recent_memory 三级检索。本文给出 skillhub 安装命令、Cursor/Claude/Codex 目录与可复现复盘模板。
Read Moredsh-agent-log: Offline Review of an Agent Work Report
`dsh-agent-log` is a DeepSeek Harness (DSH) plugin for offline restoration and generation of agent work reports. It requires no API key and reads tool calls, command execution, file read/write operations, and token consumption data from local session logs. It can be viewed via CLI, HTTP endpoints, or the right-side panel. After installation, run the `work_report({})` command to view it; in the desktop version, it must be enabled manually. The plugin runs completely offline, does not modify source data, and is suitable for reviewing and debugging agent tasks.
Read Moredsh-plugin-session-export: DeepSeek Harness Session Black Box Plugin
dsh-plugin-session-export is a debugging plugin for DeepSeek Harness (DSH) that captures comprehensive session events via bypass monitoring without interfering with Agent execution. Installation requires registering dependencies via a local link. It supports one-click generation of structured retrospective reports (in Markdown or HTML format) that include session overviews, tool invocation statistics, and error summaries. Featuring sensitive data masking, real-time statistics, hot configuration updates, and a web-based dashboard, the plugin is ideal for debugging and retrospective analysis scenarios for Agent developers.
Read More