keepkeys

Solid

Store named secrets outside the conversation and use them through KeepKeys without returning plaintext to the agent. Use when a user asks to add, list, remove, or use a secret with KeepKeys.

AI & Automation 0 stars 0 forks Updated 2 days ago Apache-2.0

Install

View on GitHub

Quality Score: 78/100

Stars 20%
0
Recency 20%
100
Frontmatter 20%
70
Documentation 15%
100
Issue Health 10%
80
License 10%
100
Description 5%
100

Skill Content

# KeepKeys KeepKeys gives the agent **use** of a local secret, not its plaintext value. ## Non-negotiable boundary - Never ask the user to paste, type, dictate, attach, or expose a secret in chat. - Never request a plaintext secret from KeepKeys; no such tool exists. - Never read or inspect the clipboard. The native KeepKeys helper may read it only after the user explicitly presses **Paste & Store**. - Never fall back to an environment file, shell profile, command argument, plugin configuration, log, or transcript. - Treat friendly names and variable names as sensitive metadata. Mention only what the current task needs. ## Store 1. If the provider and intended use are unclear, ask only for that non-secret context. Never ask for the credential value. 2. Research the credential before opening KeepKeys. Prefer AI-readable official documentation such as `llms.txt`, OpenAPI specifications, plain-text API references, or official SDK documentation. If none is available, use the provider's official human-readable credential or API documentation. Use one to three official HTTPS links and do not invent URLs. 3. Choose the short friendly name, uppercase environment-variable name, useful one-line description, provider, and documentation links. These are agent-owned metadata; never ask the user to type or edit them. 4. Call `keepkeys_store` with only those non-secret metadata fields. 5. KeepKeys shows the metadata as read-only context. The user copies the ...

Details

Author
barnlabs
Repository
barnlabs/keepkeys
Created
2 days ago
Last Updated
2 days ago
Language
Python
License
Apache-2.0

Similar Skills

Semantically similar based on skill content — not just same category