Brave Api Setup

ReviewAudited by ClawScan on May 1, 2026.

Overview

This skill is coherent for Brave Search setup, but it will reveal a Brave API key from your logged-in browser session and save it into OpenClaw configuration.

This appears safe for its stated purpose if you really want OpenClaw configured for Brave Search. Before using it, be comfortable letting the agent reveal your Brave API key in the logged-in browser and write it into ~/.openclaw/openclaw.json; protect or rotate the key if it is exposed unexpectedly.

Findings (3)

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.

Medium

#ASI03: Identity and Privilege Abuse

What this means

The agent will handle a real API credential for the user's Brave Search account; misuse of that key could affect account API usage.

Why it was flagged

The workflow uses the user's authenticated browser session to reveal and retrieve a Brave Search API key.

Skill content

"User must be logged in (openclaw browser profile)" ... "The result field contains the exact API key."

Recommendation

Run this only when you intentionally want Brave Search configured, verify the browser is on the Brave API dashboard, and rotate the key if you believe it was exposed.

Low

#ASI02: Tool Misuse and Exploitation

What this means

OpenClaw's persistent web search configuration will be changed, which is expected for the setup task but should be user-intended.

Why it was flagged

The included helper directly modifies the user's OpenClaw configuration file to store the Brave API key.

Skill content

const CONFIG_PATH = path.join(process.env.HOME, '.openclaw', 'openclaw.json'); ... config.tools.web.search.apiKey = apiKey; ... fs.writeFileSync(CONFIG_PATH, JSON.stringify(config, null, 2));

Recommendation

Confirm you want this config change before running the helper, and consider backing up or reviewing ~/.openclaw/openclaw.json afterward.

Low

#ASI09: Human-Agent Trust Exploitation

What this means

Users should not interpret the statement as a guarantee that the secret is invisible to the agent runtime or local command handling.

Why it was flagged

The wording is aimed at avoiding transcription errors, but the workflow still handles the exact key in tool results and passes it into a command or config patch.

Skill content

"The key never passes through LLM text generation." ... "node <skill_dir>/scripts/apply-api-key.js \"<extracted-key>\""

Recommendation

Treat the Brave API key as sensitive; prefer a safer secret-handling path such as stdin or a trusted config patch mechanism when available.