xurl
Fail
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill instructions provide a command to download and execute a shell script directly from a remote GitHub repository without prior verification. Evidence:
curl -fsSL https://raw.githubusercontent.com/xdevplatform/xurl/main/install.sh | bashinSKILL.md.- [EXTERNAL_DOWNLOADS]: The skill relies on fetching software and configuration from multiple external sources, including GitHub, NPM, Homebrew, and Go registries.- [COMMAND_EXECUTION]: The skill involves executing thexurlbinary with various flags and arguments, including raw API requests that can accept arbitrary JSON bodies and custom headers.- [DATA_EXFILTRATION]: The skill interacts with the~/.xurlfile, which is a known path for storing sensitive API keys, secrets, and OAuth 2.0 refresh tokens. While the skill includes instructions to the agent not to expose this file, the capability to access it exists.- [INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface where untrusted data from social media could influence agent behavior. * Ingestion points: Data fromxurl search,xurl timeline,xurl mentions, andxurl dmsinSKILL.md. * Boundary markers: None present. * Capability inventory: Shell command execution, network requests, and media file uploads via thexurltool inSKILL.md. * Sanitization: None identified.
Recommendations
- HIGH: Downloads and executes remote code from: https://raw.githubusercontent.com/xdevplatform/xurl/main/install.sh - DO NOT USE without thorough review
- AI detected serious security threats
Audit Metadata