Returns authoritative-looking AI-generated responses that are not based on actual user data.
npx skills add https://github.com/ThuHuynh1438/pm-kit --skill ask-thuhuynh1438
Please help me install this Agent Skill. Skill: ask Source: https://github.com/ThuHuynh1438/pm-kit/tree/main/.claude/skills/ask Command: npx skills add https://github.com/ThuHuynh1438/pm-kit --skill ask-thuhuynh1438
Fast access to answers stored in your vault by querying decisions, blockers, and docs with smart search.
Ask a question like 'what did we decide about auth?' to receive a grounded answer with sources from your vault.
High-intent search queries and answers about installing and using this skill.
Search your vault by asking a natural language question to retrieve grounded answers from decisions, blockers, and docs. The Skill queries directories like decisions/ and docs/ and returns a concise answer with citations to source files.
QMD hybrid search combines BM25, vector, and rerank methods to deliver fast, contextual answers from your vault. It activates when available and an index is present, otherwise falling back to a standard vault grep search.
Yes, you can get vault answers without the QMD index because the Skill automatically falls back to vault grep. It searches across decisions/, blockers/, docs/, and 01-index/ folders and indicates the search mode used in the result.
Query your vault blockers by asking who is blocked on a project to get a fast, grounded answer. The Skill searches the blockers/ directory and returns concise findings with direct citations to the source files.
The best way to get cited sources is querying through this Skill, which returns grounded results with citations to vault locations. It searches decisions, blockers, and docs using QMD or grep and references the exact source files.
Vault search operates with a default collection to query decisions, blockers, and docs across specified directories. This ensures the grep fallback and QMD hybrid search target the correct folders for grounded answers.
steipeteIndex local files and search them with BM25, vector, and hybrid queries.
thecloudtipsSynthesize sourced answers with inline citations and confidence ratings.
iagocavalcanteSearch local Markdown notes using BM25 keyword and semantic similarity search.
swarm-ai-safetyQuery run-index.yaml and vault claims for experiment metadata by tag, date, type, or claim.
