grok-search

PassAudited by ClawScan on May 10, 2026.

Overview

The visible artifacts show a coherent xAI/Grok search wrapper; it uses an xAI API key and sends user-selected requests to xAI, but no hidden exfiltration or destructive behavior is evident.

Before installing, make sure you are comfortable giving this skill access to an xAI API key and sending queries, prompts, and any selected images to xAI. Use a dedicated/revocable key where possible, avoid sending sensitive content unless acceptable, verify search citations, and run the self-test only if you intend to make real API calls.

Findings (4)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

It is harder to independently verify where the bundled scripts came from.

Why it was flagged

The registry metadata does not identify a source repository or package provenance. No remote installer is shown, so this is a provenance note rather than evidence of malicious behavior.

Skill content
Source: unknown
Recommendation

Install only if you trust the registry publisher or have reviewed the bundled scripts you will run.

What this means

Your search queries and prompts can be sent to xAI, and returned web/X content should be treated as untrusted until verified.

Why it was flagged

The skill intentionally gives the agent a wrapper around xAI's external search tools. This is purpose-aligned, but queries leave the local environment and may consume provider quota.

Skill content
Search the web or X/Twitter using xAI Grok server-side tools (web_search, x_search) via the xAI Responses API.
Recommendation

Use the skill for intended research tasks, avoid sensitive queries unless you accept xAI processing them, and verify important citations.

What this means

The agent can use the configured xAI key, which may consume quota or incur charges on that account.

Why it was flagged

The script deliberately obtains an xAI API key from the environment or Clawdbot config, including a related skill-entry fallback, to authenticate provider requests.

Skill content
process.env.XAI_API_KEY || j?.env?.XAI_API_KEY || j?.env?.vars?.XAI_API_KEY || j?.skills?.entries?.["grok-search"]?.apiKey || j?.skills?.entries?.["search-x"]?.apiKey
Recommendation

Use a dedicated, revocable xAI API key if possible and monitor provider usage.

What this means

If you run the self-test, it will execute local Node scripts and use the configured xAI API key for several test calls.

Why it was flagged

The self-test script spawns Node child processes to run bundled scripts and make real API calls. It is not shown as an automatic install step.

Skill content
const p = spawn(process.execPath, args, { cwd: __dirname, env: process.env, stdio: ["ignore", "pipe", "pipe"] });
Recommendation

Run selftest.mjs only intentionally and only when you are comfortable spending xAI API quota.