with one click
longClaw
longClaw contains 16 collected skills from jinglong92, with repository-level occupation coverage and site-owned skill detail pages.
Skills in this repository
Coding Agent 完整工作流——自主理解 codebase、制定修改计划、执行修改、运行验证、失败换路,交付可验证的代码改动。ENGINEER 路由且任务涉及代码修改时触发。
longClaw 运行时体检与诊断。用于检查 skill 不触发、hook 不生效、DEV LOG 异常、state.db 记账异常、memory 检索退化、配置迁移不完整等问题。Use when the user says 体检、自检、诊断、doctor、status、排查 longClaw、skill 不触发、hook 有问题、sidecar 异常、DEV LOG 乱了.
Generate publication-quality academic diagrams, methodology figures, architecture illustrations, and statistical plots from text descriptions using the PaperBanana multi-agent AI pipeline. Also evaluate diagram quality against reference images. Use when: (1) user asks to generate, create, or make a research diagram, methodology figure, system architecture illustration, pipeline diagram, or framework figure, (2) user asks to create a statistical plot, bar chart, or data visualization from CSV/JSON data, (3) user asks to evaluate or score a generated diagram against a reference, (4) user asks to refine or improve a previously generated diagram. NOT for: analyzing existing images, general image generation (non-academic), or chart/graph discussions without explicit generation intent.
记忆增强陪伴——BRO/SIS 回复前自动 spawn MemoryAgent 检索近期记忆,让陪伴有"记得你"的感觉。无需用户触发,CTRL 路由到 BRO/SIS 时自动激活。
面向实现闭环的工程 workflow。用于把“需求/思路/方案”转成可交付改动,强调验收标准、最小改动、立即验证、明确回滚点。冲突优先级最低:若请求同时包含“接入审计”或“修 bug/继续查”等排障闭环意图,应让位于 skill-safety-audit 或 research-execution-protocol。
研究型工程执行协议。用于复杂实现、排障、代码修改、配置修复与实验验证。强调先证据后判断、先验证后宣称完成、失败后系统性换路,不改变全局人格。冲突优先级次高:当请求包含“修 bug/继续查/验证闭环”与“交付落地”混合意图时,优先于 research-build。
分析岗位 JD,给出匹配度评级、核心要求拆解、简历叙事建议和投递优先级。
论文深度解读——Fact/Inference 分离、方法论、SOTA 对比、Reviewer#2 批判、工业落地评估、面试压缩版。
学术论文配图自动生成——输入方法描述文本和图注,通过 Retriever→Planner→Stylist→Visualizer→Critic 五代理流水线生成发表级配图。需本地安装 PaperBanana(https://github.com/dwzhu-pku/PaperBanana)。
主动心跳——由 cron 定期触发 heartbeat-agent 巡检,发现有价值信息写入 heartbeat-state.json;用户下次开口时 CTRL 读取并在回复开头呈现待处理事项。
Trigger and execute session compression plus post-compression handoff for next-session continuity. Use when user asks to compress long conversations, enforce compression thresholds, or standardize how compressed summaries are written into daily memory and MEMORY.md for later recall.
审计外部 SKILL.md / agent 仓库 / 自动化脚本的接入风险。用于判断某个 skill 是否应原样引入、拆解学习或禁止接入 longClaw。冲突优先级最高:当请求同时包含“接入审计”与“实现/排障”意图时,先触发本技能。
并发多源深度调研——把复杂问题拆解为多个并发 SearchAgent,各自独立搜索后由 CTRL 汇总,输出带来源的综合报告。适合需要多角度、多来源的复杂调研任务。
核查最新信息——价格/资讯/技术动态/政策变化,多源交叉验证,区分确定信息和推断信息,注明时效。
对公开网页/论文执行只读检索与证据摘录,返回 exact query、URL、逐字片段、段落位置与简短解释。
Bootstrap and migrate an OpenClaw workspace from single-agent to beginner-friendly multi-agent mode with visible routing. Use when user asks to create/adapt/sync multi-agent setup, add role definitions, enforce routing line output, or quickly replicate a proven multi-agent architecture.