vultr/openclaw
1
0
Fork 0
mirror of https://github.com/openclaw/openclaw.git synced 2026-05-04 01:00:29 +00:00
Code Issues Packages Projects Releases Wiki Activity

Tools: add x_search via xAI Responses

Browse Source
This commit is contained in:
huntharo
2026-03-27 12:41:02 -04:00
committed by Peter Steinberger
parent 5ed8ee6832
commit 38e4b77e60
34 changed files with 1111 additions and 40 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
docs/providers/xai.md
View File

@@ -27,6 +27,9 @@ openclaw onboard --auth-choice xai-api-key
}
```
OpenClaw now uses the xAI Responses API as the bundled xAI transport. The same
`XAI_API_KEY` can also power Grok-backed `web_search` and first-class `x_search`.
## Current bundled model catalog
OpenClaw now includes these xAI model families out of the box:
@@ -52,9 +55,9 @@ openclaw config set tools.web.search.provider grok
- Auth is API-key only today. There is no xAI OAuth/device-code flow in OpenClaw yet.
- `grok-4.20-multi-agent-experimental-beta-0304` is not supported on the normal xAI provider path because it requires a different upstream API surface than the standard OpenClaw xAI transport.
- Native xAI server-side tools such as `x_search` and `code_execution` are not yet first-class model-provider features in the bundled plugin.
## Notes
- OpenClaw applies xAI-specific tool-schema and tool-call compatibility fixes automatically on the shared runner path.
- `web_search` and `x_search` are exposed as OpenClaw tools. OpenClaw enables the specific xAI built-in it needs inside each tool request instead of attaching both search tools to every chat turn.
- For the broader provider overview, see [Model providers](/providers/index).

4
docs/tools/grok-search.md
View File

@@ -12,6 +12,9 @@ OpenClaw supports Grok as a `web_search` provider, using xAI web-grounded
responses to produce AI-synthesized answers backed by live search results
with citations.
The same `XAI_API_KEY` can also power the built-in `x_search` tool for X
(formerly Twitter) post search.
## Get an API key
<Steps>
@@ -69,4 +72,5 @@ Provider-specific filters are not currently supported.
## Related
- [Web Search overview](/tools/web) -- all providers and auto-detection
- [x_search in Web Search](/tools/web#x_search) -- first-class X search via xAI
- [Gemini Search](/tools/gemini-search) -- AI-synthesized answers via Google grounding

28
docs/tools/index.md
View File

@@ -52,19 +52,19 @@ OpenClaw has three layers that work together:
These tools ship with OpenClaw and are available without installing any plugins:
| Tool | What it does | Page |
| ---------------------------- | -------------------------------------------------------- | --------------------------------- |
| `exec` / `process` | Run shell commands, manage background processes | [Exec](/tools/exec) |
| `browser` | Control a Chromium browser (navigate, click, screenshot) | [Browser](/tools/browser) |
| `web_search` / `web_fetch` | Search the web, fetch page content | [Web](/tools/web) |
| `read` / `write` / `edit` | File I/O in the workspace | |
| `apply_patch` | Multi-hunk file patches | [Apply Patch](/tools/apply-patch) |
| `message` | Send messages across all channels | [Agent Send](/tools/agent-send) |
| `canvas` | Drive node Canvas (present, eval, snapshot) | |
| `nodes` | Discover and target paired devices | |
| `cron` / `gateway` | Manage scheduled jobs, restart gateway | |
| `image` / `image_generate` | Analyze or generate images | |
| `sessions_*` / `agents_list` | Session management, sub-agents | [Sub-agents](/tools/subagents) |
| Tool | What it does | Page |
| --------------------------------------- | -------------------------------------------------------- | --------------------------------- |
| `exec` / `process` | Run shell commands, manage background processes | [Exec](/tools/exec) |
| `browser` | Control a Chromium browser (navigate, click, screenshot) | [Browser](/tools/browser) |
| `web_search` / `x_search` / `web_fetch` | Search the web, search X posts, fetch page content | [Web](/tools/web) |
| `read` / `write` / `edit` | File I/O in the workspace | |
| `apply_patch` | Multi-hunk file patches | [Apply Patch](/tools/apply-patch) |
| `message` | Send messages across all channels | [Agent Send](/tools/agent-send) |
| `canvas` | Drive node Canvas (present, eval, snapshot) | |
| `nodes` | Discover and target paired devices | |
| `cron` / `gateway` | Manage scheduled jobs, restart gateway | |
| `image` / `image_generate` | Analyze or generate images | |
| `sessions_*` / `agents_list` | Session management, sub-agents | [Sub-agents](/tools/subagents) |
For image work, use `image` for analysis and `image_generate` for generation or editing. If you target `openai/*`, `google/*`, `fal/*`, or another non-default image provider, configure that provider's auth/API key first.
@@ -115,7 +115,7 @@ Use `group:*` shorthands in allow/deny lists:
| `group:fs` | read, write, edit, apply_patch |
| `group:sessions` | sessions_list, sessions_history, sessions_send, sessions_spawn, sessions_yield, subagents, session_status |
| `group:memory` | memory_search, memory_get |
| `group:web` | web_search, web_fetch |
| `group:web` | web_search, x_search, web_fetch |
| `group:ui` | browser, canvas |
| `group:automation` | cron, gateway |
| `group:messaging` | message |

80
docs/tools/web.md
View File

@@ -1,11 +1,12 @@
---
summary: "web_search tool -- search the web with Brave, Firecrawl, Gemini, Grok, Kimi, Perplexity, or Tavily"
read_when:
- You want to enable or configure web_search
- You need to choose a search provider
- You want to understand auto-detection and provider fallback
title: "Web Search"
sidebarTitle: "Web Search"
summary: "web_search, x_search, and web_fetch -- search the web, search X posts, or fetch page content"
read_when:
- You want to enable or configure web_search
- You want to enable or configure x_search
- You need to choose a search provider
- You want to understand auto-detection and provider fallback
---
# Web Search
@@ -13,6 +14,10 @@ sidebarTitle: "Web Search"
The `web_search` tool searches the web using your configured provider and
returns results. Results are cached by query for 15 minutes (configurable).
OpenClaw also includes `x_search` for X (formerly Twitter) posts and
`web_fetch` for lightweight URL fetching. In this phase, `web_fetch` stays
local while `web_search` and `x_search` can use xAI Responses under the hood.
<Info>
`web_search` is a lightweight HTTP tool, not browser automation. For
JS-heavy sites or logins, use the [Web Browser](/tools/browser). For
@@ -40,6 +45,12 @@ returns results. Results are cached by query for 15 minutes (configurable).
await web_search({ query: "OpenClaw plugin SDK" });
```
For X posts, use:
```javascript
await x_search({ query: "dinner recipes" });
```
</Step>
</Steps>
@@ -136,6 +147,9 @@ Provider-specific config (API keys, base URLs, modes) lives under
`plugins.entries.<plugin>.config.webSearch.*`. See the provider pages for
examples.
For `x_search`, configure `tools.web.x_search.*` directly. It uses the same
`XAI_API_KEY` fallback as Grok web search.
### Storing API keys
<Tabs>
@@ -195,6 +209,55 @@ examples.
-- use their dedicated tools for advanced options.
</Warning>
## x_search
`x_search` queries X (formerly Twitter) posts using xAI and returns
AI-synthesized answers with citations. It accepts natural-language queries and
optional structured filters. OpenClaw only enables the built-in xAI `x_search`
tool on the request that serves this tool call.
### x_search config
```json5
{
tools: {
web: {
x_search: {
enabled: true,
apiKey: "xai-...", // optional if XAI_API_KEY is set
model: "grok-4-1-fast-non-reasoning",
inlineCitations: false,
maxTurns: 2,
timeoutSeconds: 30,
cacheTtlMinutes: 15,
},
},
},
}
```
### x_search parameters
| Parameter | Description |
| ---------------------------- | ------------------------------------------------------ |
| `query` | Search query (required) |
| `allowed_x_handles` | Restrict results to specific X handles |
| `excluded_x_handles` | Exclude specific X handles |
| `from_date` | Only include posts on or after this date (YYYY-MM-DD) |
| `to_date` | Only include posts on or before this date (YYYY-MM-DD) |
| `enable_image_understanding` | Let xAI inspect images attached to matching posts |
| `enable_video_understanding` | Let xAI inspect videos attached to matching posts |
### x_search example
```javascript
await x_search({
query: "dinner recipes",
allowed_x_handles: ["nytfood"],
from_date: "2026-03-01",
});
```
## Examples
```javascript
@@ -223,13 +286,13 @@ await web_search({
## Tool profiles
If you use tool profiles or allowlists, add `web_search` or `group:web`:
If you use tool profiles or allowlists, add `web_search`, `x_search`, or `group:web`:
```json5
{
tools: {
allow: ["web_search"],
// or: allow: ["group:web"] (includes both web_search and web_fetch)
allow: ["web_search", "x_search"],
// or: allow: ["group:web"] (includes web_search, x_search, and web_fetch)
},
}
```
@@ -238,3 +301,4 @@ If you use tool profiles or allowlists, add `web_search` or `group:web`:
- [Web Fetch](/tools/web-fetch) -- fetch a URL and extract readable content
- [Web Browser](/tools/browser) -- full browser automation for JS-heavy sites
- [Grok Search](/tools/grok-search) -- Grok as the `web_search` provider