chrome-extension-research
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a bundled script,
scripts/crawlora.sh, which acts as a wrapper for the vendor's REST API. The script is implemented with security controls including a restricted path whitelist and checks to prevent command-line parameter injection. - [EXTERNAL_DOWNLOADS]: Network requests are performed against
api.crawlora.net. This domain is the official API endpoint for the skill's author, crawlora-org, and is required for the skill's intended operation. - [INDIRECT_PROMPT_INJECTION]: The skill processes external data including extension descriptions and user reviews. Ingestion points are primarily the
/search,/item, and/reviewsAPI endpoints. While boundary markers are not explicitly provided in the prompt instructions, the skill handles this data through a helper script that provides sanitization for query parameters. - [SAFE]: A thorough security review of the instructions and bundled scripts found no malicious patterns. The script notably blocks curl's ability to read local files via query parameters, mitigating a common exfiltration vector.
Audit Metadata