microsim-screen-capture
Fail
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: HIGHCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION] (HIGH): The script
capture-screenshot.shexecutes Google Chrome/Chromium with flags that disable the Same-Origin Policy (SOP) and permit file-to-file access. This represents high-risk command execution as it intentionally compromises the security boundaries of the browser sandbox. - [DATA_EXFILTRATION] (HIGH): The specific browser configuration enables a 'Local File Disclosure' (LFD) attack vector. Malicious JavaScript within a processed
main.htmlfile can programmatically read local host files (e.g., SSH keys, environmental variables) and transmit them to a remote domain. - [PROMPT_INJECTION] (LOW): The skill is vulnerable to indirect prompt injection via the untrusted
main.htmlfile. - Ingestion points: The
main.htmlfile in the directory provided by the user. - Boundary markers: Absent; there are no delimiters or warnings to prevent the browser from executing malicious embedded scripts.
- Capability inventory: Full browser execution with local file read permissions and network access.
- Sanitization: None; the HTML content is rendered directly without validation or script filtering.
Recommendations
- AI detected serious security threats
Audit Metadata