vellum-self-knowledge

Answer questions about Vellum architecture and configuration via CLI, docs, and source code.

2|2|Updated May 13, 2026
One-click install
npx skills add https://github.com/autonomy-cloud/kairos-interface --skill vellum-self-knowledge-autonomy-cloud
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: vellum-self-knowledge
Source: https://github.com/autonomy-cloud/kairos-interface/tree/main/skills/vellum-self-knowledge
Command: npx skills add https://github.com/autonomy-cloud/kairos-interface --skill vellum-self-knowledge-autonomy-cloud

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps users quickly understand the assistant’s current architecture, capabilities, and configuration without guessing or relying on outdated assumptions.

Core Features & Use Cases

  • Live answers from the assistant CLI: Pulls current model/provider, full configuration, installed skills, platform connection state, auth/identity, trust rules, credentials, API keys, MCP servers, autonomy tier, watchers, token usage/costs, and version—using the CLI as the source of truth.
  • Conceptual grounding from the Vellum docs site: Fetches authoritative explanations for “what is/how/why” questions from the Vellum documentation.
  • Deep implementation lookups when needed: Uses the Vellum assistant open-source codebase (and the matching release tag) to answer questions the docs and CLI cannot fully cover.

Quick Start

Ask: "What model is my Vellum assistant running, what skills are installed, and what’s in my current configuration?"

Frequently Asked Questions about vellum-self-knowledge

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

FAQPage Schema
How do I check what model my Vellum assistant is currently running?

Check the currently running model by querying the assistant CLI for live runtime state. The CLI acts as the source of truth for active provider, configuration, and connection status, ensuring accurate runtime diagnostics without relying on outdated assumptions.

How do I list installed skills and inspect active configuration in Vellum?

List installed skills and inspect active configuration by running an introspection query against the assistant CLI. This retrieves full configuration details, active trust rules, credentials, and API keys directly from the live runtime environment.

What is the best way to find conceptual documentation for Vellum architecture questions?

The best way to find conceptual documentation is routing architecture explainer queries to the Vellum docs site. This fetches authoritative explanations for what, how, and why questions about the assistant's architecture and capabilities directly from official documentation.

Can I look up deep implementation details when the Vellum docs and CLI do not cover a specific question?

Yes, you can look up deep implementation details using the Vellum assistant open-source codebase. Source code lookup matches the current release tag to answer questions that CLI introspection and documentation retrieval cannot fully resolve.

Does the Vellum assistant answer configuration questions from memory if the live sources are unavailable?

No, the assistant applies a fail-closed policy and does not answer from memory if live sources are unavailable. This strict resolution order prioritizes the assistant CLI, Vellum docs site, and source code lookup to ensure capability discovery and configuration answers remain accurate.