Knowledge Agent

This turns your claude-mem observations into queryable AI sessions. Build a filtered corpus from your observation history (say, all hook-related decisions or last month's bugfixes), prime it to load the knowledge into a chat session, then ask questions conversationally. It's useful when you want focused answers from past work rather than scanning through everything manually. The workflow is straightforward: build with filters like project, types, concepts, or date ranges, prime once, then query repeatedly. The session remembers context across follow-ups. Keep corpora focused for better results, and rebuild plus reprime when you've added new observations worth including.

skills.sh installs: 6K
GitHub stars: 115.5K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add thedotmack/claude-mem --skill knowledge-agent --agent claude-code

Installs into .claude/skills of the current project.

Files

Knowledge Agent

Build and query AI-powered knowledge bases from claude-mem observations.

What Are Knowledge Agents?

Knowledge agents are filtered corpora of observations compiled into a conversational AI session. Build a corpus from your observation history, prime it (loads the knowledge into an AI session), then ask it questions conversationally.

Think of them as custom "brains": "everything about hooks", "all decisions from the last month", "all bugfixes for the worker service".

Workflow

Step 1: Build a corpus

build_corpus name="hooks-expertise" description="Everything about the hooks lifecycle" project="claude-mem" concepts="hooks" limit=500

Filter options:

project — filter by project name
types — comma-separated: decision, bugfix, feature, refactor, discovery, change
concepts — comma-separated concept tags
files — comma-separated file paths (prefix match)
query — semantic search query
dateStart / dateEnd — ISO date range
limit — max observations (default 500)