🔍

Kagi

Verified

by michaelasper

Gives your agent Kagi web search and FastGPT answers through Kagi's API. Bundles three dependency-free Python wrappers: kagi_search.py for a ranked result list, kagi_fastgpt.py for a short answer with cited reference URLs, and a shared kagi_client.py. Bring your own Kagi API token from https://kagi.com/settings/api and expose it as KAGI_API_TOKEN. Useful when you want Kagi's result quality instead of the default engine, or a fallback when another search backend is rate-limited.

searchkagiweb-searchresearchfastgpt
View on GitHub

Kagi (API)

Use the bundled Python scripts to call Kagi’s API from the OpenClaw host.

Quick start

1) Create a token in https://kagi.com/settings/api

2) Export it for your shell/session:

export KAGI_API_TOKEN='…'

3) Run a search:

python3 scripts/kagi_search.py "haaps glass" --limit 10 --json

4) Or ask FastGPT (LLM + web search):

python3 scripts/kagi_fastgpt.py "Summarize the latest Haaps glass mentions" --json

Tasks

1) Web search (Kagi Search API)

Use when you need a normal ranked list of results (URLs/titles/snippets).

Command:

python3 scripts/kagi_search.py "<query>" [--limit N] [--json]

Notes:

  • Defaults to printing a readable digest; use --json for raw API output.
  • The script automatically sets Authorization: Bot <token>.

2) Answer/summarize with citations (FastGPT)

Use when you want a short answer grounded in web results, including reference URLs.

Command:

python3 scripts/kagi_fastgpt.py "<question>" [--cache true|false] [--json]

3) Using Kagi as a drop-in for web_search

If Brave Search is rate-limited (429) or you want better results:

  • Use scripts/kagi_search.py to fetch results
  • Then use the main agent model to synthesize / summarize based on the returned URLs/snippets

Files

  • API reference snippets: references/kagi-api.md
  • Python client + CLIs: scripts/kagi_client.py, scripts/kagi_search.py, scripts/kagi_fastgpt.py