# `mix pixir.smoke.web_search`
[🔗](https://github.com/Ranvier-Technologies/pixir/blob/main/lib/mix/tasks/pixir.smoke.web_search.ex#L1)

Probes Pixir's OpenAI Responses hosted `web_search` path.

This is a manual, opt-in, real-network smoke. It verifies request shaping and whether
the active backend/model emits Provider-hosted web search evidence. It does not use
Pixir local browser automation and does not treat search output as Pixir's source of
truth.

Usage:

    mix pixir.smoke.web_search --dry-run --json
    mix pixir.smoke.web_search --json
    mix pixir.smoke.web_search --model gpt-5.5 --reasoning-effort low --json
    mix pixir.smoke.web_search --search-context-size low --json
    mix pixir.smoke.web_search --help

Options:

  * `--model MODEL` - Provider model. Default: `gpt-5.5`.
  * `--reasoning-effort EFFORT` - one of `low`, `medium`, `high`, `xhigh`.
    Default: `low`.
  * `--search-context-size SIZE` - hosted web search context size: `low`, `medium`,
    or `high`. Default: `low`.
  * `--query TEXT` - search prompt. Default asks for OpenAI's web search docs.
  * `--dry-run` - validate and print the planned request shape without auth/network.
  * `--json` - print machine-readable evidence or errors.
  * `--help` - print this help and exit.

---

*Consult [api-reference.md](api-reference.md) for complete listing*