Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/hainrixz/skill-vault/claude-mdgit clone --depth 1 https://github.com/Hainrixz/skill-vaultWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/hainrixz/skill-vault/claude-md)<a href="https://agentmods.dev/instructions/hainrixz/skill-vault/claude-md"><img src="https://agentmods.dev/badge/instructions/hainrixz/skill-vault/claude-md.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.02594 | $0.02594 |
| Opus 5 | $0.01297 | $0.01297 |
| Sonnet 5 | $0.00519 | $0.00519 |
| Haiku 4.5 | $0.00259 | $0.00259 |
Grade E, and why
skill-vault CLAUDE.md scanned grade E with 3 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
- `curl | bash`, `wget | sh`, or any pipe-to-shell execution Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
- `rm -rf` or recursive deletion commands Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- `curl | bash`, `wget | sh`, or any pipe-to-shell execution How it starts
The opening of the file, as written. The whole thing — 227 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill Vault — Vault Master Agent
You are the Vault Master, a skill organizer, security analyst, and recommender. Your sole purpose is managing the user's personal skill library.
This project is a Skill Vault — a zero-dependency tool for organizing, analyzing, and searching Claude Code skills. Users find skills across the internet (GitHub, skills.sh, skillsmp.com, npm packages) and bring them here for safekeeping, security analysis, and easy retrieval.
Project Structure
skill-vault/
├── CLAUDE.md # You are here — agent instructions
├── catalog.md # Master index of all saved skills
├── security-rubric.md # Detailed security analysis reference
├── categories/ # Skill entries organized by category
│ ├── <category>/
│ │ ├── _category.md # Category description + local index
│ │ └── <skill>.md # Individual skill entries
└── templates/ # Templates for new entries and categories
├── skill-entry.md
└── category-readme.md
Key Files
catalog.md— The master index. Always read this first when searching or listing skills.security-rubric.md— Detailed security analysis reference. Consult when analyzing skills.templates/skill-entry.md— Template for new skill entries. Follow this format exactly.templates/category-readme.md— Template for new category folders.
Commands
/vault-add — Add a Skill to the Vault
Accepts: GitHub URL, raw SKILL.md content pasted by the user, a skill name to search for, or a file path.
Workflow:
-
Retrieve the skill content:
- If a GitHub URL: use WebFetch to retrieve the raw SKILL.md content. If the URL points to a repo (not a specific file), look for
SKILL.mdin the root orskills/directory. - If raw content is pasted: parse it directly.
- If a skill name: search skills.sh or GitHub for the skill, fetch its SKILL.md.
- If a local file path: read the file.
- If a GitHub URL: use WebFetch to retrieve the raw SKILL.md content. If the URL points to a repo (not a specific file), look for
-
Check for duplicates:
- Search
catalog.mdfor an existing skill with the same name. - If a match is found, inform the user and ask whether to: update the existing entry, save as a new entry with a different name, or cancel.
- If no match, continue.
- Search
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago First seen · 227 lines · 2,594 tokens per session scan E 7af6226c64e7
skill-vault CLAUDE.md is an instructions file published in the GitHub repository Hainrixz/skill-vault (67 stars, last pushed 5mo ago), licensed MIT. It adds 2,594 tokens to every session, about $0.0130 per session on Opus 5. A static security scan graded it E with 3 findings (downloads and executes remote code, recursive force delete, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
agent-2-agent CLAUDE.md
Instructions for CtriXin/agent-2-agent, covering a2a — ai agent 快速上下文, 这是什么, 文件职责, 关键设计决策 and 修改注意事项.
FableClaudeMDForOpus CLAUDE.md
Claude Code instructions for altafino/FableClaudeMDForOpus, covering iron rules, project and hard stops.
AmbyKit CLAUDE.md
Claude Code instructions for ambystechcom/AmbyKit, covering claude code — notes for the ambykit repo and claude-specific.
skillsync-mcp CLAUDE.md
Instructions for adityasugandhi/skillsync-mcp, covering claude.md, project overview, development commands, architecture and key files.
MCPs copilot-instructions.md
Copilot instructions for AceDataCloud/MCPs, covering copilot sync instructions for acedatacloud mcps, repository structure, source of truth, what to sync and rules.
agent-memory AGENTS.md
AGENTS.md instructions for ac0033/agent-memory, covering agents.md — agent-memory 仓库协作规范, 三条架构红线(不可违反), 可信根清单(d6 的枚举,禁止 agent 自修改), 工程约定 and 架构地图与当前状态(m8 完成).