knowledge-agent

1# Knowledge Agent
2 
3Build and query AI-powered knowledge bases from claude-mem observations.
4 
5## What Are Knowledge Agents?
6 
7Knowledge 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.
8 
9Think of them as custom "brains": "everything about hooks", "all decisions from the last month", "all bugfixes for the worker service".
10 
11## Workflow
12 
13### Step 1: Build a corpus
14 
15```text
16build_corpus name="hooks-expertise" description="Everything about the hooks lifecycle" project="claude-mem" concepts="hooks" limit=500
17```
18 
19Filter options:
20- `project` — filter by project name
21- `types` — comma-separated: decision, bugfix, feature, refactor, discovery, change
22- `concepts` — comma-separated concept tags
23- `files` — comma-separated file paths (prefix match)
24- `query` — semantic search query
25- `dateStart` / `dateEnd` — ISO date range
26- `limit` — max observations (default 500)
27 
28### Step 2: Prime the corpus
29 
30```text
31prime_corpus name="hooks-expertise"
32```
33 
34This creates an AI session loaded with all the corpus knowledge. Takes a moment for large corpora.
35 
36### Step 3: Query
37 
38```text
39query_corpus name="hooks-expertise" question="What are the 5 lifecycle hooks and when does each fire?"
40```
41 
42The knowledge agent answers from its corpus. Follow-up questions maintain context.
43 
44### Step 4: List corpora
45 
46```text
47list_corpora
48```
49…