spectra-ask

Retrieve grounded answers from openspec documents via spectra CLI vector search.

33|8|Updated Jan 13, 2026
One-click install
npx skills add https://github.com/PsychQuant/che-ical-mcp --skill spectra-ask-psychquant
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: spectra-ask
Source: https://github.com/PsychQuant/che-ical-mcp/tree/main/.agents/skills/spectra-ask
Command: npx skills add https://github.com/PsychQuant/che-ical-mcp --skill spectra-ask-psychquant

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enables querying the openspec/documents repository and returning answers grounded strictly in those documents, preventing reliance on external knowledge or assumptions.

Core Features & Use Cases

  • Grounded document-based Q&A: answers are derived exclusively from openspec sources.
  • Multilingual query support: understands Chinese, English, and Japanese inputs via the spectra pipeline.
  • Transparent sourcing: returns references to the matched openspec files used to generate each answer.

Quick Start

Ask a question after the spectra-ask trigger to retrieve grounded answers from openspec documents.

Frequently Asked Questions about spectra-ask

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

FAQPage Schema
How do I get grounded answers from openspec documents?

You can retrieve grounded answers by triggering a query that uses vector search to read matched openspec files, strictly sourcing responses from those documents and returning references to prevent reliance on external knowledge.

Can I ask documentation questions in Chinese or Japanese?

Yes, multilingual query support allows you to ask documentation questions in Chinese, English, and Japanese. The pipeline processes these inputs to perform vector search and retrieve grounded answers from openspec documents.

What is the best way to prevent a Q&A model from relying on external knowledge?

The best way to prevent reliance on external knowledge is to use a grounded Q&A approach that strictly uses openspec sources for sourcing. This ensures answers are derived exclusively from the documents, returning transparent references to matched files.

Does the spectra pipeline show which openspec files were used for an answer?

Yes, the spectra pipeline provides transparent sourcing by returning references to the matched openspec files used to generate each answer. This ensures you can verify the exact documents sourced for the grounded response.

Are there limitations to grounded document-based Q&A from openspec sources?

A key limitation of grounded document-based Q&A is that it strictly sources answers exclusively from openspec documents. It will not provide responses or make assumptions based on external knowledge outside of the indexed repository.