heterocat-skills
CLAUDE.md4 starsChanged 8 months agoApache-2.0View on GitHub
This is HeteroCat Skills - a Claude Code skills library containing agent skills for content creation, AI image generation, financial analysis, research, and more. Skills are located in .claude/skills/. Skills are organized with category prefixes: Output directories: - .skill files → dist/ - .zip files → zip_dist/ Each skill directory follows this structure: The SKILL.md file must include YAML frontmatter: All API keys and credentials are stored in .env at the repository root. Key variables include: Skills can be installed…
- Reads credentials
In this repository47 other files it gives its agents.
- .claude/skills/f-fund-advisor/CLAUDE.mdCLAUDE.md
- .claude/skills/t-supabase/CLAUDE.mdCLAUDE.md
- agent-browserSkill
- find-skillsSkill
- skill-creatorSkill
- fund-advisorSkill
- fundamentalsSkill
- fund-assistantSkill
- fund-screenerSkill
- eastmoney_fin_dataSkill
- eastmoney_fin_searchSkill
- f-mx-select-stockSkill
- mx_self_selectSkill
- eastmoney_stock_simulatorSkill
- baoyu-article-illustratorSkill
- xhs-note-creatorSkill
- auto-wechatSkill
- auto-xSkill
- doubao-image-genSkill
- doubao-videoSkill
- minimaxSkill
- nano-banana2Skill
- remotion-best-practicesSkill
- suno-musicSkill
- whisper-subtitleSkill
- arxivSkill
- bar-chart-race-generatorSkill
- perfect-PPTmakerSkill
- tech-news-dailySkill
- x-apiSkill
- yt-dlp-downloaderSkill
- brainstormingSkill
- dispatching-parallel-agentsSkill
- executing-plansSkill
- finishing-a-development-branchSkill
- frontend-designSkill
- receiving-code-reviewSkill
- requesting-code-reviewSkill
- subagent-driven-developmentSkill
- supabase-postgres-best-practicesSkill
- systematic-debuggingSkill
- test-driven-developmentSkill
- using-git-worktreesSkill
- using-superpowersSkill
- verification-before-completionSkill
- writing-plansSkill
- writing-skillsSkill
The file
We split this CLAUDE.md into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 86
All API keys and credentials are stored in .env at the repository root.
- Distinctive: line 127
When updating external skills, verify the source repository for changes.
≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Repository Overview
This is HeteroCat Skills - a Claude Code skills library containing agent skills for content creation, AI image generation, financial analysis, research, and more. Skills are located in .claude/skills/.
Skill Naming Conventions
Skills are organized with category prefixes:
| Prefix | Category | Examples |
a- | Agent tools | a-agent-browser, a-find-skills, a-skill-creator |
f- | Financial | f-fund-assistant, f-mx-data, f-fundamentals |
m- | Media/Content | m-auto-redbook, m-auto-wechat |
o- | Output/Creation | o-generate-video, o-nano-banana2, o-minimax |
p- | Productivity | p-arxiv, p-tech-news-daily, p-PPTmaker |
t- | Team/Workflow | t-brainstorming, t-writing-plans, t-systematic-debugging |
Common Commands
Package All Skills
# Package as .skill files (Claude Code native format)
python package_skills.py
# Package as .zip files (universal format)
python package_zip.pyOutput directories:
.skillfiles →dist/.zipfiles →zip_dist/
Create a New Skill
# Initialize new skill structure
python .claude/skills/a-skill-creator/scripts/init_skill.py my-skill --path ./.claude/skills
# Package a single skill
python .claude/skills/a-skill-creator/scripts/package_skill.py .claude/skills/my-skillSkill Development
# Validate skill structure
python .claude/skills/a-skill-creator/scripts/quick_validate.py .claude/skills/my-skill
# Run skill evaluation (creates test scenarios)
python .claude/skills/a-skill-creator/scripts/run_eval.py .claude/skills/my-skillSkill Structure
Each skill directory follows this structure:
.claude/skills/{skill-name}/
├── SKILL.md # Required: skill definition with frontmatter
├── scripts/ # Optional: executable scripts
│ └── *.py, *.js, *.sh
├── references/ # Optional: documentation, schemas
│ └── *.md
├── assets/ # Optional: images, templates
│ └── *
└── agents/ # Optional: subagent definitions
└── *.mdThe SKILL.md file must include YAML frontmatter:
---
name: skill-name
description: "Description of when to use this skill"
---Environment Variables
All API keys and credentials are stored in .env at the repository root. Key variables include:
| Variable | Purpose |
OPENROUTER_API_KEY | AI image generation (nano-banana2) |
MINIMAX_API_KEY | Voice synthesis & music generation |
ARK_API_KEY | Doubao video generation |
X_BEARER_TOKEN | X/Twitter read access |
X_API_KEY, X_API_SECRET, X_ACCESS_TOKEN, X_ACCESS_TOKEN_SECRET | X/Twitter OAuth 1.0a |
XHS_COOKIE | Xiaohongshu auto-posting |
MX_APIKEY | Dongfang Wealth financial data |
DMX_API_KEY | Suno music generation |
WECHAT_APP_ID, WECHAT_APP_SECRET | WeChat Official Account posting |
Key Files
| File | Purpose |
.env | Environment variables (API keys) |
skills-lock.json | Tracks external skills with git hashes |
.claude/settings.local.json | Claude Code local settings |
package_skills.py | Build script for .skill format |
package_zip.py | Build script for .zip format |
Installing Skills
Skills can be installed in Claude Code via:
/skills install /path/to/skill-name.skillExternal Skills
Some skills are sourced from external GitHub repos and tracked in skills-lock.json:
- fund-advisor → realqiyan/fund-advisor
- fund-assistant → xer97/fund-assistant
- fund-screener → sososun/mutual-fund-skills
- fundamentals → staskh/trading_skills
- agent-browser → vercel-labs/agent-browser
When updating external skills, verify the source repository for changes.
Language
默认使用中文回复用户。
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

