← all publishers

ligenjian001-ai

@ligenjian001-ai source repo

Report this profile Claim this profile

• 20 skills
• 0 followers
• 15 hours ago last updated

1. ▌
   Ag Archive · ligenjian001-ai bundle

   Conversation Journal protocol for preserving decision context, error patterns, and user intent across AG conversations.
   0 installs
2. ▌
   Browser Use · ligenjian001-ai bundle

   Browser automation via 3 layers: AG browser_subagent (built-in), HybridAgent (browser-use + CDP keyboard), and noVNC fallback. Use this skill for web tasks requiring navigation, text input into SPAs, cookie sharing, or human-in-the-loop.
   0 installs
3. ▌
   Deploy Docs · ligenjian001-ai

   Build Sphinx HTML/PDF documentation and deploy to a remote server via rsync + nginx. Trigger: '发布文档', 'deploy docs', 'publish docs', '部署文档'.
   0 installs
4. ▌
   Jenkins Ops · ligenjian001-ai bundle

   Manage Jenkins CI/CD jobs — trigger, monitor, configure, and troubleshoot. Use for any Jenkins-related operation.
   0 installs
5. ▌
   Daily Report · ligenjian001-ai bundle

   生成个人日报/周报。自动搜集AG对话、Panel Discussion、Task Delegate数据，按项目分组总结，附AI改进建议和明日事项初稿。触发词：生成日报、daily report、今天做了什么、工作总结。
   0 installs
6. ▌
   Infra Intake · ligenjian001-ai

   MANDATORY when user sends a GitHub issue URL or says '处理 issue'. AG confirms the problem, then delegates design and implementation to CC through multi-round dialogue. Covers hft_build C++/Python SDK, scripts, CMake.
   0 installs
7. ▌
   Deep Analysis · ligenjian001-ai bundle

   Multi-dimensional system analysis — AG elicits user intent through dialogue, dispatches focused investigators, and surfaces decision points with tradeoffs for user judgment.
   0 installs
8. ▌
   Infra Request · ligenjian001-ai bundle

   MANDATORY when filing infra requests. Covers Python API-first deliverable priority, GitHub Issues submission, idempotency guard, post-submit verification, project-aware labeling, and Self-Verification Guide.
   0 installs
9. ▌
   Skill Creator · ligenjian001-ai bundle

   Standards and workflow for creating high-quality agent skills. Defines the documentation paradigm, quality checklist, and skill structure conventions.
   0 installs
10. ▌
    System Design · ligenjian001-ai bundle

    Detailed system design — multi-perspective sketch, user-driven convergence at each decision point, real-world validation, and implementation blueprint with direct execution path.
    0 installs
11. ▌
    Task Delegate · ligenjian001-ai bundle

    Delegate tasks to any backend (CC, Codex, Gemini, DeepSeek). AG acts as orchestrator — prepares prompt, launches backend in tmux, monitors, and verifies results.
    0 installs
12. ▌
    Tmux Protocol · ligenjian001-ai bundle

    MANDATORY terminal command protocol via tmux with PS1-based automatic markers. MUST be read before ANY terminal operation.
    0 installs
13. ▌
    Jupyter Session · ligenjian001-ai bundle

    Persistent Jupyter kernel sessions for data analysis. Execute Python code, manage kernels, handle image outputs via Jupyter Server REST API. Use when user needs data exploration, plotting, or interactive computation.
    0 installs
14. ▌
    Ag Lag Diagnosis · ligenjian001-ai bundle

    Diagnose and fix Antigravity IDE Agent Manager lag/stalling. Covers streamAgentStateUpdates polling storms, oversized conversation .pb files, VERY LONG TASK renderer blocking, and memory leaks.
    0 installs
15. ▌
    API Design Audit · ligenjian001-ai bundle

    Audit API design proposals against user's design preferences. Evaluates 7 checkpoints (raw data preservation, metric justification, core concept, data constraints, minimal state, additive design, market transparency). Trigger: 'audit this API design', '审计API设计', 'review this proposal'.
    0 installs
16. ▌
    Bug Investigation · ligenjian001-ai bundle

    Multi-agent bug investigation — dispatch bug to two independent Codex analysts (code + logic) for root cause analysis, AG synthesizes the verdict.
    0 installs
17. ▌
    Agent Native Product · ligenjian001-ai bundle

    Transform an SDK/CLI/API project's interface to be agent-friendly. Assess maturity, generate documentation, audit quality. Trigger: 'agent化改造', 'make API agent-friendly', 'improve agent usability'.
    0 installs
18. ▌
    Concept Model Review · ligenjian001-ai bundle

    Review a product/SDK/framework's core conceptual model — assess completeness, orthogonality, layering, and design the opening path. Trigger: '概念模型审计', 'concept review', '产品架构评估', 'API 概念梳理'.
    0 installs
19. ▌
    Agent Panel Discussion · ligenjian001-ai bundle

    Multi-agent panel discussion — Codex subagents debate a topic with preset stances across multiple rounds, producing an HTML report with academic-style references.
    0 installs
20. ▌
    Product Maturity Audit · ligenjian001-ai

    Shell-based product maturity checklist for strategy platform SDKs. Runs concrete commands with binary PASS/FAIL outcomes. Designed to avoid the bootstrapping paradox of Python-based checkers.
    0 installs