Skip to main content
在 Manus 中运行任何 Skill
一键导入
hu-qi
GitHub 创作者资料

hu-qi

按仓库查看 21 个 GitHub 仓库中的 38 个已收集 skills。

已收集 skills
38
仓库
21
更新
2026-07-15
这里展示前 8 个仓库;完整仓库列表在下方继续。
仓库浏览

仓库与代表性 skills

using-wisedev
项目管理专家

enforce wisedev workflow discipline before product-delivery tasks. use when the user asks for requirement clarification, requirement specifications, design documents, openapi yaml contracts, vue plus mock prototypes, or any request that may cross stages. require the assistant to check the wisedev skill suite first, route multi-stage tasks through wisedev-orchestrator first, and map clear single-stage requests to the correct wisedev child skill.

2026-04-22
wisedev-orchestrator
项目管理专家

orchestrate staged delivery artifact generation from short requirements, meeting notes, transcripts, and mixed materials. use when the user wants to turn ambiguous chinese product or business input into structured outputs such as requirement briefs, requirement specifications, design documents, openapi yaml contracts, or vue plus mock prototypes. act as the mandatory entry skill for multi-stage work: identify the current stage first, normalize context first, choose one child skill first, and require stage handoff artifacts before any downstream stage.

2026-04-22
wisedev-vue-mock-prototype
软件开发工程师

generate vue-based frontend prototype plans and code scaffolds with local mock services from chinese requirement, design, and openapi artifacts. use when the user needs a modern vue 3 plus vite plus typescript plus pinia plus router prototype, with realistic mock data, workflow states, and local run instructions for complex enterprise business demonstrations.

2026-04-22
wisedev-prototype-design
软件开发工程师

recommend contrasting prototype design directions, generate theme token packages, and critique visual outputs for chinese enterprise-facing prototypes. use when the user needs design direction, visual style comparison, color schemes, prototype themes, or design review before or after vue prototype generation.

2026-04-21
wisedev-design-doc
管理分析师

generate chinese high-level and detailed design documents from requirement specifications, requirement briefs, or structured business input. use when the user needs architecture decomposition, module boundaries, role permissions, workflows, state transitions, data entities, interfaces, and exception handling documented in markdown in a formal enterprise style.

2026-04-19
wisedev-meeting-to-spec
管理分析师

analyze meeting notes, transcript text, and mixed chinese discussion materials and convert them into structured requirement inputs. use when the user provides a transcript, a meeting summary, or scattered discussion text and wants requirement clarification, feature extraction, role and scenario analysis, unresolved question tracking, or a draft requirement brief.

2026-04-19
wisedev-openapi-contract
软件开发工程师

generate openapi yaml contracts from requirement and design artifacts. use when the user needs chinese business requirements converted into mock-ready api paths, request and response schemas, error models, pagination, security, and example payloads in a single self-contained openapi yaml file.

2026-04-19
wisedev-requirement-spec
管理分析师

write structured chinese requirement specifications from rough needs, clarified requirement briefs, transcripts, or mixed materials. use when the user needs a formal markdown requirement document with scope, actors, scenarios, functional requirements, non-functional requirements, acceptance criteria, and traceable identifiers in a government or enterprise documentation style.

2026-04-19
harbor-environment-design
软件开发工程师

Build or repair a deterministic Harbor task environment with explicit resources, network policy, health checks, and safe solver-visible boundaries.

2026-06-24
harbor-oracle-verifier
软件质量保证分析师与测试员

Implement a legitimate Harbor Oracle solution and validate its behavior against an independent verifier without weakening the task contract.

2026-06-24
harbor-redteam-verifier
软件质量保证分析师与测试员

Red-team a Harbor verifier for answer leakage, reward hacking, shared-container tampering, artifact manipulation, and semantic under-testing.

2026-06-24
harbor-rollout-calibration
软件开发工程师

Calibrate a frozen tb-hard task using Harbor Oracle and real-agent rollouts, then distinguish authentic difficulty from ambiguity, flakiness, leakage, or infrastructure failure.

2026-06-24
harbor-task-create
软件开发工程师

Create or evolve a canonical Harbor task inside a stable tb-hard case directory. Use after the design brief names observable acceptance criteria and at least three difficulty dimensions.

2026-06-24
tb-hard-release-audit
软件开发工程师

Audit a calibrated Harbor task for deterministic purchaser export, leakage, evidence completeness, and release readiness.

2026-06-24
tb2-pattern-research
软件开发工程师

Research Harbor and Terminal-Bench 2 patterns to design an original tb-hard task. Use before approving a design brief or when checking whether a candidate is too shallow or derivative.

2026-06-24
generate-app-spec
管理分析师

Use this skill whenever a user wants to generate a specification document for an autonomous AI coding agent — not write the code itself, but create the spec that an agent will execute. Trigger for: "generate app spec", "create app_spec.txt", "spec out my app idea", "帮我生成 app spec", "生成规格文档", "帮我准备 app_spec.txt", "写一份 PRD 给 agent 用", "为自主编码 Agent 写规格". Also trigger when the user says they want to build something "with an autonomous coding agent", "with claude-agent", "with autonomous-coding-demo", or "用自主编码 Agent 构建" and needs the spec file first. Produces a complete XML app_spec.txt through a short interview covering app idea, tech stack, features, DB schema, API endpoints, and step-by-step implementation plan. Skip for: direct code generation, scaffolding an existing project, debugging, refactoring, or writing documentation intended for human engineers rather than AI agents.

2026-05-06
project-architecture-summarizer
软件开发工程师

Summarize a software project's high-level architecture into an ARCHITECTURE.md-style document. Use this skill whenever the user wants to understand how a codebase is organized, generate an architecture overview, write an ARCHITECTURE.md, or figure out "where things live" in a project. Trigger on phrases like "architecture", "how is this project structured", "code map", "ARCHITECTURE.md", "understand this codebase", "project overview", or when the user is onboarding to an unfamiliar repo. Also trigger when the user asks about entry points, boundaries between modules, or cross-cutting concerns like logging, configuration, or error handling across the whole project.

2026-05-06
已展示 12 / 21 个仓库