he search¶

Perform semantic search over a knowledge abstract.

Synopsis¶

he search KA_PATH QUERY [OPTIONS]

Arguments¶

Argument	Description
`KA_PATH`	Path to knowledge abstract directory
`QUERY`	Search query string

Options¶

Option	Short	Description	Default
`--top-k`	`-n`	Number of results to return	3

Description¶

Semantic search finds relevant information even when keywords don't match exactly. It uses:

Vector embeddings — Converts query and content to vectors
Similarity matching — Finds semantically similar content
Ranking — Returns most relevant results

Requires: Search index must be built. Run he build-index if needed.

Examples¶

Basic Search¶

he search ./output/ "Tesla's inventions"

Get More Results¶

he search ./output/ "electrical engineering" -n 10

Natural Language Queries¶

he search ./ka/ "What were the major achievements?"
he search ./ka/ "People who worked with Edison"
he search ./ka/ "Important dates in the timeline"

After Building Index¶

# First, ensure index exists
he build-index ./output/

# Then search
he search ./output/ "innovation"

Output Format¶

Found 3 result(s):

Result 1:
{
  "name": "Nikola Tesla",
  "type": "person",
  "description": "Serbian-American inventor, electrical engineer..."
}

Result 2:
{
  "source": "Nikola Tesla",
  "target": "Thomas Edison",
  "type": "worked_with",
  "description": "Tesla worked for Edison in 1884"
}

Result 3:
...

How It Works¶

Query Embedding — Converts your query to a vector
Index Search — Finds nearest vectors in the knowledge abstract
Result Ranking — Returns top-k most similar items

Tips for Better Search¶

Use natural language — "inventions in electrical engineering" vs "invention electrical"
Be specific — "Tesla's work on AC power" vs "Tesla work"
Try synonyms — If "inventions" doesn't work, try "discoveries"
Increase top-k — Use -n 10 for broader results

Comparison with `he talk`¶

Feature	`he search`	`he talk`
Returns	Raw entities/relations	Natural language answer
Use case	Find specific data	Get explanations
Speed	Faster	Slower (LLM generation)
Precision	Exact matches	Interpretive

Troubleshooting¶

"Index not found"¶

Build the search index:

he build-index ./output/

"No results found"¶

Try: 1. Broader query terms 2. Increase -n for more results 3. Different phrasing 4. Check he info ./output/ to verify data exists