Search memory wiki pages and return answers with citations.
npx skills add https://github.com/dbbaskette/herald --skill wiki-query-dbbaskette
Please help me install this Agent Skill. Skill: wiki-query Source: https://github.com/dbbaskette/herald/tree/main/skills/wiki-query Command: npx skills add https://github.com/dbbaskette/herald --skill wiki-query-dbbaskette
This Skill helps Herald retrieve and cite relevant information from Herald's long-term memory wiki instead of relying on model memory, ensuring grounded responses.
Use the wiki-query skill to fetch knowledge on a topic from MEMORY.md and sources/* with citations.
High-intent search queries and answers about installing and using this skill.
To recall knowledge with citations, this Skill searches memory pages by terms, loads related concepts and sources from the memory wiki, and returns answers with explicit citations preserving provenance.
Retrieving information from a memory wiki involves searching stored pages and cross-links to load relevant concepts, ensuring responses are grounded in long-term memory rather than relying on model memory alone.
Handling missing data when querying a memory wiki is managed gracefully, returning available citations and relevant pages while avoiding errors when the requested knowledge or memory pages do not exist.
Yes, you can use memory search to perform preflight checks for other skills by loading required memory pages and fetching relevant input data before executing the primary workflow.
Wiki-query supports one-shot recall workflows by surfacing stored memory pages and cross-links to answer specific topic questions while preserving the provenance of the original sources.
dbbasketteIngest local files, URLs, or pasted text into Herald's structured wiki memory.
JoturesQuery Google NotebookLM notebooks via Python and PatchScript automation with source citations.
arbisoftCentralize knowledge management across local files, MCP memory, vector stores, and Git repositories.
rukikotooQuery the econometrics RAG knowledge base to retrieve relevant document chunks.
