hermes-vault

Stores and retrieves passwords, server configs, and user preferences in categorized Markdown files.

Updated Jul 3, 2026
One-click install
npx skills add https://github.com/CHENHUI-X/toolbox --skill hermes-vault-chenhui-x
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: hermes-vault
Source: https://github.com/CHENHUI-X/toolbox/tree/main/custom-skills/productivity/hermes-vault
Command: npx skills add https://github.com/CHENHUI-X/toolbox --skill hermes-vault-chenhui-x

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Users lose track of passwords, API keys, server configurations, and personal preferences scattered across conversations and config files. This Skill acts as a persistent vault, writing and reading categorized Markdown files under ~/.hermes/vault/ so information survives across sessions without consuming memory token quotas. ## Core Features & Use Cases - Categorized Storage: Organizes data into five Markdown files covering credentials, server configs, completed tasks, preferences, and useful tips. - Write and Recall on Demand: Saves information when the user says "remember this" and retrieves it when the user asks questions like "what is my API key". - Plain-Text Permanence: Uses standard .md files that users can also open and edit directly on their machine. - Use Case: A user asks the assistant to remember a new API key with its provider, base URL, and model name; later, in a new session, the user asks for the key and the assistant reads it back from the vault file. ## Quick Start Ask the assistant to remember a piece of information such as an API key or server IP, and it will be saved into the appropriate vault file for later retrieval.

Frequently Asked Questions about hermes-vault

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I store passwords and API keys with an AI assistant?

Tell the assistant to remember the credential, and it writes the value into the categorized Markdown file ~/.hermes/vault/🔐 密码和密钥.md. The entry records the provider, API address, model name, and the config file location where the key was found.

How do I retrieve saved information from the vault?

Ask a natural question such as what a specific key or server IP is, and the assistant reads the corresponding Markdown file in ~/.hermes/vault/ and extracts the answer. No special query syntax is required.

Where are the vault files stored and can I edit them manually?

All vault data lives as plain-text Markdown files under ~/.hermes/vault/ on the local machine. Because they are standard .md files, you can open, view, and edit them directly with any text editor.

Does storing data in the vault consume memory or token quota?

No. Vault data is written to plain-text files on disk rather than kept in conversational memory, so it does not consume memory token quota. Only lightweight preference information stays in memory.

What happens if an API key is masked in the config file?

The assistant records the masked value exactly as shown along with the config file name and line number. If a full key is requested, it attempts to locate it in auth.json, environment variables, or process environ, and reports honestly if it cannot be found.