Skillv0.1.3

ClawScan security

Google Maps Api Skill · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

SuspiciousMar 17, 2026, 10:56 AM

Verdict: suspicious
Confidence: medium
Model: gpt-5-mini
Summary: The skill's code and instructions largely match its stated purpose (using BrowserAct to collect Google Maps data), but there are a few mismatches and operational risks you should understand before installing.
Guidance: This skill appears to do what it claims (call BrowserAct to scrape Google Maps), but take these precautions before installing: - Confirm you trust the BrowserAct service and that you want to provide it with an API key (the script will transmit your key to api.browseract.com). - Be aware the SKILL.md expects the agent to 'proactively trigger' the skill; if the agent can invoke skills autonomously, configure invocation policies or require explicit user confirmation to avoid unwanted mass scraping. - Ensure the runtime environment has the Python 'requests' package installed (the skill does not declare this dependency), or the script will fail. - Note the registry metadata omitted marking the API key as the primary credential; verify the skill's configuration in the registry/UI before granting secrets. - Consider legal and privacy implications of automated scraping for your use case (terms of service, regional laws). If you need stronger assurance, request the publisher's homepage/source provenance and an explicit dependency list before enabling the skill.

Review Dimensions

Purpose & Capability: okName, description, declared env var (BROWSERACT_API_KEY), and the included script all point to the same purpose: invoking BrowserAct's Google Maps workflow and returning results. The network endpoints (api.browseract.com) and the template ID in the script are consistent with the described capability.
Instruction Scope: noteSKILL.md instructs the agent to run the provided Python script, watch terminal logs, and retry once on non-auth failures — this stays within the scraping task. However the SKILL.md also emphasizes the agent should "proactively trigger" the skill; combined with autonomous invocation this could lead to repeated, automated scraping tasks if the agent is configured to invoke skills without tight user supervision.
Install Mechanism: concernThere is no install spec, which is low-risk, but the script imports the Python 'requests' package even though only 'python' is listed as a required binary. If 'requests' is not available at runtime the script will fail. The lack of an explicit dependency declaration (e.g., pip requirement) is an operational mismatch that could cause runtime errors or encourage ad-hoc installation steps by the agent.
Credentials: noteThe skill only requests a single, relevant environment variable (BROWSERACT_API_KEY) which is proportional to its purpose. Two minor issues: (1) the registry metadata lists 'Primary credential: none' despite requiring BROWSERACT_API_KEY, which is inconsistent; (2) the skill will send the API key to api.browseract.com (Authorization: Bearer ...), so users should ensure they trust that service and the key's scope/permissions.
Persistence & Privilege: okThe skill does not request persistent or privileged installation (always is false). It does not modify other skills or system config. Autonomous invocation is enabled by default but not combined with an 'always' flag or other elevated privileges.