learning-debug
Skill1 starsChanged 33 days agoMITView on GitHub
快速只读检查 Agent Learning 插件连接、当前对话绑定、项目知识目录及管理员模型、预算和阻塞原因;适用于调试接入,不用于知识浏览、发布或开启审核。
In this repository7 other files it gives its agents.
- AGENTS.mdAGENTS.md
- learning-adminSkill
- learning-browserSkill
- learning-historySkill
- learning-panelSkill
- learning-reviewSkill
- learning-workerSkill
The file
Files
Agent Learning 调试
这是输入框技能菜单的调试快捷项。调用现有 MCP 工具检查当前项目,不打开控制页面或运行管理员。
- 按当前工具 schema 调用一次
learning_status(view="summary"),提取project_id、home、host_scope、storage_schema_version、version、管理员状态及预算。旧安装尚无view参数时才使用无参调用;只有摘要缺少本次诊断必需字段或标记strings_truncated时才按需读view="full",不把缩短的路径或模型当完整身份。不回显完整响应、知识正文、会话句柄或访问码。工具缺失时报告尚未接入,不改安装或配置。 - 仅使用本轮宿主 Hook 注入的真实
context_handle调用session_context。没有句柄时仅查状态并报告“本轮 Hook 未验证/缺少句柄”,不无参调用后借用连接中的旧绑定。不查找“最新会话”、复用历史句柄或猜测身份。项目不符、句柄失效或需重连时停止绑定检查,给出对应原因。 - 简短列出:MCP 是否响应、Hook 对话是否绑定、当前项目及知识目录、管理员实际开关
admin_admission_enabled、模型模式与admin_routing.executor.model、project_usage.counted_attempts项目今日审核尝试、completed_reviews有效结论、failed_attempts失败、project_usage.remaining_calls项目剩余额度;另列global_usage所有项目合计与全局保护余量以及admission.blocking_scope,注明按 UTC 日统计与本地重置时间;缺少项目字段时报告“项目统计暂不可用”,不能把全局数当项目数;token 未上报应标“未知”。同时说明当前阻塞原因。无模型观察或路由未就绪时显示admin_routing.reason,不猜默认模型。旧安装缺少有效开关字段时,将admin_enabled标注为“策略开关,实际准入未提供”。开关打开或 CLI 可用都不等于审核器正在运行或常驻。 - 工具响应只能证明工具连通;只有本轮真实句柄成功绑定才报告本对话 Hook 已接通。不要把已有知识数量、旧事件或空知识库当成连接成功或失败依据。
- 默认仅作上述读取,不修改任务、反馈、配置或知识,不启动/重试审核、历史吸收或安装。详细知识查看转到 learning-browser;候选审批与管理员设置转到 learning-review,在用户明确要求时再打开。
用用户当前语言给出短状态清单和一个最相关的下一步,不生成调试文档。
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

