commands.md

Commands & Examples

Common Flags

• --api-key: Override EXA_API_KEY.
• --pretty: Indented JSON output.
• --save <file>: Save JSON to a file.

Search

exa search --query "2025 LLM agent benchmarks" --type fast --text --pretty

Options:

• --type: auto|neural|keyword|fast (see https://docs.exa.ai/reference/how-exa-search-works)
• Include/exclude domains; date bounds; include/exclude text terms. Docs: https://docs.exa.ai/reference/search

Contents

exa contents https://example.com --text --livecrawl preferred

Options:

• --text, --highlights
• --livecrawl: always|preferred|fallback|never (https://docs.exa.ai/reference/livecrawling-contents) Docs: https://docs.exa.ai/reference/get-contents

Find Similar

exa find-similar --url https://arxiv.org/abs/2307.06435 --num-results 5

Docs: https://docs.exa.ai/reference/find-similar-links

Answer

exa answer --query "What changed in Exa 2.0?" --pretty

Docs: https://docs.exa.ai/reference/answer

Research

Create:

exa research start --instructions @examples/research_instructions.md \
  --schema @examples/research_schema.json --model exa-research-fast

Get:

exa research get --id <researchId> [--events]

List:

exa research list --limit 10

Poll (presets: fast=10s, balanced=30s, pro=40s):

exa research poll --id <researchId> --preset balanced

Stream (JSON-lines):

exa research stream --id <researchId> | jq .

Context (Exa Code)

exa context query --query "pandas groupby examples" --tokensNum dynamic

Documentation (Exa Code)

• Context (Exa Code): https://docs.exa.ai/reference/context

Examples

• See examples/ for runnable scripts:
  • examples/search_examples.sh: search filters and inline contents
  • examples/contents_examples.sh: contents options (text/highlights/summary/metadata)
  • examples/find_similar_examples.sh: filters plus inline contents
  • examples/answer_examples.sh: structured and streaming answers (JSON-lines)
  • examples/research_examples.sh: start/poll/stream/get/list flows
  • examples/context_example.sh: Exa Code context queries
  • examples/stream_consumer.sh: pretty-prints JSON-lines
  • examples/agents_python.py: Python agent helpers
  • examples/pipeline_search_contents_answer.py: Search → Contents → Answer pipeline
  • examples/research_stream_json.py: Research streaming (JSON-lines)
  • examples/context_rag_snippet.py: Context for RAG prompts
  • examples/research_instructions.md: Research instructions template
  • examples/research_schema.json: Research schema template

Exa API Documentation

• Search: https://docs.exa.ai/reference/search
• Contents: https://docs.exa.ai/reference/get-contents
• Find Similar: https://docs.exa.ai/reference/find-similar-links
• Answer: https://docs.exa.ai/reference/answer
• Research:
  • Create: https://docs.exa.ai/reference/research/create-a-task
  • Get/Stream: https://docs.exa.ai/reference/research/get-a-task
  • List: https://docs.exa.ai/reference/research/list-tasks
• Context (Exa Code): https://docs.exa.ai/reference/context
• Livecrawling Contents: https://docs.exa.ai/reference/livecrawling-contents

Exa Research

Documentation (Exa Research)

• Create: https://docs.exa.ai/reference/research/create-a-task
• Get/Stream: https://docs.exa.ai/reference/research/get-a-task
• List: https://docs.exa.ai/reference/research/list-tasks