spectra-ask

Answer questions using only content from openspec documents with vector search.

Updated Apr 10, 2026
One-click install
npx skills add https://github.com/kaochenlong/handshake --skill spectra-ask-kaochenlong
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: spectra-ask
Source: https://github.com/kaochenlong/handshake/tree/main/.claude/skills/spectra-ask
Command: npx skills add https://github.com/kaochenlong/handshake --skill spectra-ask-kaochenlong

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Answers are grounded in openspec/documents and must not rely on general knowledge or training data. If documents do not contain the answer, indicate so.

Core Features & Use Cases

  • Grounded QA: provide answers derived only from openspec content.
  • Vector search with cross-language support to locate relevant documents, prioritizing main specs over archive entries.
  • Use case: quickly answer questions about spectra capabilities using only documentation.

Quick Start

Ask a question after /spectra-ask and receive a grounded answer drawn only 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 answer questions using only my project documentation?

To answer questions using only project documentation, query the openspec knowledge base to receive grounded responses sourced solely from those documents, ensuring no external training data is used.

Can I search across multiple languages in an openspec knowledge base?

Yes, you can search across multiple languages in an openspec knowledge base. The vector search mechanism supports cross-language queries to locate relevant documents while prioritizing main specs over archived entries.

What happens if my documentation does not contain the answer to a query?

If the documentation does not contain the answer to a query, the system provides a clear fallback message indicating the content is unavailable, strictly maintaining a read-only and document-grounded QA workflow.

How do I retrieve answers from openspec documents without hallucination?

To retrieve answers without hallucination, the QA workflow grounds responses exclusively in openspec documents and returns sourced references, completely ignoring general knowledge or training data.

Does the spectra-ask search prioritize current specs over archived documents?

Yes, the spectra-ask search prioritizes current main specs over archived documents. This ensures the retrieved answers and sourced references reflect the most relevant and active project documentation.

What is the best way to query openspec documents for project knowledge?

The best way to query openspec documents is through a read-only QA workflow that uses vector search to locate relevant content, returning sourced references and fallback messages when information is missing.