venomwise/skill-factory
Instructions file CodexOpenCode
AGENTS.md instructions for venomwise/skill-factory, covering repository guidelines, project structure & module organization, build, test, and development commands, build db-explorer from source and build exa-search from source.
venomwise/skill-factory
Instructions file
Claude Code instructions for venomwise/skill-factory: This repository keeps a single source of truth for agent and contributor guidance: AGENTS.md.
venomwise/skill-factory
Skill Claude CodeCodex
A planning method for turning a large or abstract software idea into a local project direction and a prioritised backlog. A backlog is a list of work still to be done, ordered by importance and dependencies.
venomwise/skill-factory
Skill Claude CodeCodex
A guided process for turning one unclear software request into a written design and acceptance criteria. It uses evidence from the project, step-by-step questions, and comparisons between possible approaches.
venomwise/skill-factory
Skill Claude CodeCodex
An independent, evidence-based review process for code changes, commits, or local pull-request differences.
venomwise/skill-factory
Skill Claude CodeCodex
Convert HTTP interface/controller code to curl command and standard HTTP message format (raw HTTP/1.1 request) examples. Use when given source code (Spring MVC, FastAPI, Express, NestJS, Gin, etc.) to generate ready-to-run API test commands. Triggered by requests like "生成 curl 请求", "转成 http 格式", "给我这个接口的请求示例", or "how…
venomwise/skill-factory
Skill Claude CodeCodex
Read-only CLI for exploring PostgreSQL, MySQL, and SQLite databases. Use this skill whenever a task needs reliable database facts, including schema details, sample records, read-only query results, or code-model verification against actual database structure. Do not use for writes, migrations, data fixes, bulk…
venomwise/skill-factory
Skill Claude CodeCodex
A skill for applying design-review findings after confirming each design decision with the user. It also handles direct fixes and updates the design documentation and readiness status.
venomwise/skill-factory
Skill Claude CodeCodex
A separate, evidence-based review process for a file named design.md, which usually describes a software design.
venomwise/skill-factory
Skill Claude CodeCodex
Neural web search for official documentation, API references, and structured source retrieval with text extraction. Use for precise, source-first research when you need high-quality results or extracted page content. For breaking news or real-time sentiment, use grok-search instead.
venomwise/skill-factory
Skill Claude CodeCodex
Generate git commit messages and submit commits. Use when preparing commit messages, deciding split commits, or running git commit.
venomwise/skill-factory
Skill Claude CodeCodex
Create and manage git worktrees for isolated parallel work. Use this skill whenever the user mentions worktrees, wants to work on multiple branches simultaneously, needs to test changes without affecting their main working directory, or asks about creating/managing/removing worktrees. Also trigger when users want to…
venomwise/skill-factory
Skill Claude CodeCodex
Real-time web research and live synthesis with sources. Use when the question depends on freshness, community chatter, X/Twitter dynamics, breaking updates, or broad multi-source summaries. Prefer this over source-first search when you need fast situational awareness. Prefer exa-search instead for official docs, API…
venomwise/skill-factory
Skill Claude CodeCodex
Fetch Hacker News stories, comments, and users. Use when the user says "Hacker News", "HN", "Y Combinator", "tech news digest", "top stories on HN", or asks to pull/aggregate tech news from HN.
venomwise/skill-factory
Skill Claude CodeCodex
Generate High-Level Design (HLD) — concise, visual overview for human review. Transforms detailed design.md (LLD) into scannable HLD with diagrams/tables (300-400 lines, 50% of source). Use for "write HLD", "概要设计", "高层设计", "generate tech design", "帮我写个技术设计", "生成设计文档". Input: design.md (from clarifying skill) or git…
venomwise/skill-factory
Skill Claude CodeCodex
A tool guide for turning Markdown—a plain-text document format—into a standalone HTML web page. The output can include navigation, summaries, highlighted key points, and document checks.
venomwise/skill-factory
Skill Claude CodeCodex
Apply Charlie Munger's mental models (多元思维模型 + 人类误判心理学 25 种心理倾向) to aid decision-making. Use when the user asks for decision support, wants to analyze a choice/investment/problem using Munger's framework, or says "芒格思维", "逆向思考", "用芒格的方法分析", "避免犯错", "决策检查". Also triggered when the user describes a dilemma, trade-off…
venomwise/skill-factory
Skill Claude CodeCodex
A process for revising a PRD, or product requirements document, after a review has identified problems or suggestions.
venomwise/skill-factory
Skill Claude CodeCodex
Analyze PRD documents for logic gaps, boundary issues, and technical feasibility before development starts. Focus on business logic closure and architectural reasonability, not implementation details.
venomwise/skill-factory
Skill Claude CodeCodex
A pre-development review for a product requirements document, or PRD—a written description of what a product should do. It checks the business process, missing edge cases, and whether the proposed technical approach is feasible.
venomwise/skill-factory
Skill Claude CodeCodex
A price-checking tool for products sold by major Chinese online marketplaces, including JD, Taobao, Tmall, and Pinduoduo. It gathers current price information without requiring a marketplace login.
venomwise/skill-factory
Skill Claude CodeCodex
Guide for creating and optimizing AI agent skills. Use when the user wants to create a new skill, improve an existing skill, review skill quality, or understand skill authoring best practices. Covers skill structure, content guidelines, common patterns, and anti-patterns.
venomwise/skill-factory
Skill Claude CodeCodex
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
venomwise/skill-factory
Agent
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.