- Skills
- 114
- 仓库
- 1
- 职业覆盖
- 0
- 场景覆盖
- 0
- Stars
- ★ 1
zaurakworks 的 Skills 114
skill-maintenance
创建、审计、修正、拆分、升级、迁移或退役 Skill,或其合同/入口已漂移时,先恢复行为、授权与预算,再同步正文、调用者、版本、生成物和验证。普通业务维护、纯格式修正、承载位置未定的纠正不触发。Use for explicit Skill maintenance, not carrier choice.
skill-appraisal
判定一个 Skill 组该不该进当前装配、归哪些事项、与谁重叠,或对已判定过的组按节拍复核。三种进入:首次判定新候选、补判早已在用但从未判定过的组、按失效条件复核。判定单位是组不是单个 Skill;产出必须留下失效条件与下次最少复核步骤。用于外部来源、他人所写、疑似重复或从未被判定过的组。从零设计新 Skill 用 agent-skill-design;已定下后的改造、迁移、退役用 skill-maintenance;只要外部标准与版本事实用 capability-lifecycle;对象是 knowledge 用 knowledge-maintenance。本 Skill 只产出判定,不安装、不写配置、不改 profile。Appraise a Skill group, or recheck an adopted one. Judgement only, never installs.
self-improvement
当用户指出 Agent 漂移、误解、重复犯错,要求把任务经验固化为系统改进,或讨论只增概念却不减关键未知、形成决定或可检验资产时,只暂停依赖被推翻假设的路径,重锚原问题、持久记录纠正、诊断原因,并在授权内改进入口提示词或 Skill。最小实验不能验收产品交付;普通一次性错误、偏好或任务内修复不触发完整自我优化。Use for reported drift, misunderstanding, repeated mistakes, durable Agent improvement, or abstraction loops without decisions or testable assets. Freeze only affected paths and avoid a full detour for ordinary one-off corrections.
resource-observability
当用户查询 Codex/Claude 账户额度、重置时间、Codex 重置券、单 Session Token,或高成本/多 Agent 工作要决定启动、并行、降级、延后或停止时,按需读取 Orca 账户快照与固定 ccusage 回执,区分状态、来源、可信度和未知值。不要用于低风险任务、轮询、自动调度/消费权益、账单推断或长期保存快照。Use for account capacity, reset times, credits, session tokens, or consequential resource decisions. Skip routine work, polling, scheduling, automatic credit use, billing inference, and durable snapshot storage.
orchestrated-collaboration
当用户明确要求多 Agent/多 Session/跨 Provider 协作、任务已授权委派,或活动 Session 发生共享写入碰撞时,建立目标来源、排他所有权、可追踪交付、独立验收与综合;按共享资源和 Issue 子树确定唯一协调者,只冻结重叠或受推翻假设影响的路径。协调者参与高价值多交付件能力的拆分或设计时,由未参与者复核父目标;使用已选后端,选 Orca 时动态读 orchestration 指南。不要因复杂、额度或空闲 Agent 擅自并行,也不把试用变成长期依赖。Use for authorized multi-agent work, cross-session delegation, or shared-write collisions. Establish exclusive ownership, traceable delivery, independent acceptance, and bounded synthesis; add parent-goal review when required and avoid invented coordination or permanent platform commitments.
knowledge-maintenance
用于多来源调研、可重复实验、会影响权威/Agent 配置/重要决定的研究,或用户要求复用、复核、更新当前知识时:先找已认可知识和失效条件,只补变化、冲突与缺口,再经价值门和可信门更新。不用于低成本一次性事实、原始材料/研发过程留存、私域结构设计、定时全量重查或旧仓导入。Use before multi-source research, repeatable experiments, consequential authority/configuration decisions, or explicit knowledge reuse. Skip cheap one-off facts, raw retention, private architecture, scheduled full re-research, and bulk legacy imports.
grilling
仅在用户直接要求 grilling/盘问/压力测试,或明确接受建议后,用结构化问题压力测试计划、决定或想法;复杂性、关键词或 Agent 偏好不构成同意。Use structured questions to stress-test a plan, decision, or idea only on direct request or explicit acceptance; complexity or keywords are not consent.
pr-integration
用于验收、修复、更新或合并 GitHub PR:证据绑定当前 head,区分 Draft、审查、required/optional checks、冲突、merge state 及人/bot/App 反馈;本段独占受门控的 Draft → ready。分支按干净、fetch、rebase、验证、lease 链维护;仅获明确授权才合并,只向 issue-workflow 返回事实而不判生命周期。Use to integrate a PR against its current head, with precise review/CI state, the sole guarded Draft-to-ready action, safe branch maintenance, and authorized merging; return facts without lifecycle decisions.
operating-ledger-maintenance
当 Agent 系统有需跨 Session 保留的诉求、候选、决定、计划/实验、交付、验收或证据变化,或用户要求维护经营总账时,恢复权威、授权与 GitHub,分开维护执行、诉求和证据状态;只投影当前、明确、未消费的负责人动作,消费后清除旧等待、写明下一责任人与动作并回读。不要用于普通任务、轮询、自动派发、未授权写入或历史迁移。Use for durable Agent-system ledger changes. Keep execution, need, and evidence separate; project only current explicit unconsumed owner actions, clear consumed waits, name the next owner/action, and re-read remote state. Skip routine work, polling, auto-dispatch, unauthorized writes, and migration.
objective-to-issues
当长程目标已对齐、GitHub 写入已授权,且持久父/子 Issue 图能改善并行交付或恢复时,建立边界清楚、可独立交付和远端核验的任务图,写明依赖、所有权、交付合同与决定门;不用于普通单 PR、未解决目标或未授权写入。Use for an aligned, authorized long-horizon objective that benefits from a persistent parent/child Issue graph. Do not use for ordinary one-PR work, unresolved goals, or unapproved GitHub writes.
issue-workflow
用于从 GitHub 推进一棵 Issue 子树、续接明确父级缺口或消费负责人短回复:以可独立核验交付物判定叶子、协调父级切片,解析决定前核验可信负责人;仅当前、明确、未消费的动作才等待负责人,关闭前登记知识候选。本 Skill 独占生命周期判定;没有具体 Issue 不扫队列,也不绕过合并、产品决定或写入所有权。Use to drive one Issue subtree and bind owner replies after trusted-principal verification. It alone routes verifiable deliverables and decides lifecycle state; do not scan queues, bypass authorization, or retain a permanent coordinator role.
issue-delivery
当已就绪 GitHub Issue 要形成可独立核验交付物时,从远端恢复范围、依赖、授权、成功条件、交付形态和所有权,按合同交付 Draft PR 或自足证据评论,并记录实际运行、未运行与缺口。已有授权不重问;不建任务图,不审查/合并自己的 PR,也不判生命周期。Use to deliver one ready Issue as its contracted Draft PR or evidence comment. Restore remote authority and ownership, validate proportionately, and return facts without task-graph, merge, or lifecycle decisions.
issue-contract-compaction
当 GitHub Issue 正文已不能代表当前合同,或准备派发/交接时,取得正文排他所有权,以可重读的远端快照保存原文,写前再读远端并收敛正文;保留评论,不改变范围、决定或授权,不用陈旧本地草稿覆盖远端。Use when an Issue body needs compaction for handoff. Require exclusive ownership, a durable remote snapshot, and a final re-read; preserve comments and never overwrite newer state or invent authority.
adaptive-problem-solving
用于问题含糊、关键路径选择、波次/里程碑反思、高成本或难回退动作、范围/成本变化、停滞、恢复/交接/验收/长期收口,或检查方向、方法、ROI、模型、上下文、工具、环境与 Agent 组合。恢复原问题和主瓶颈,比较普通路径与方法后组合、换路或退出,区分证据来源、实验与完整交付,并核对父目标和能力回退;跳过低风险易回退步骤。方法的执行流程若已由装配内其他 Skill 承载(如 bmad 的 brainstorming、advanced-elicitation、forge-idea、party-mode、deep-recon),本 Skill 只负责归型、进入与退出条件、成本与停止,不重复执行。Use for ambiguity, consequential choices, milestones, stalls, handoffs, acceptance, or execution review. Restore the problem and bottleneck, compare methods, and verify evidence, parent goals, and regressions; skip low-risk reversible work.
review-skill
暂无简介
implementation-skill
暂无简介
bmad-validate-prd
Deprecated — forwards to bmad-prd (validate intent).
bmad-ux
Plan UX patterns and design specifications. Use when the user says "lets create UX design" or "create UX specifications" or "help me plan the UX"