xurl
Fail
Audited by Gen Agent Trust Hub on Mar 14, 2026
Risk Level: CRITICALREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill promotes an installation method that pipes a remote shell script directly into the bash interpreter:
curl -fsSL https://raw.githubusercontent.com/xdevplatform/xurl/main/install.sh | bash. This allows for unverified execution of arbitrary code from a non-whitelisted third-party source. - [EXTERNAL_DOWNLOADS]: The skill instructions specify downloading software from several external sources including GitHub, NPM (@xdevplatform/xurl), and Homebrew (xdevplatform/tap/xurl), none of which are on the provided list of trusted vendors.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it retrieves and processes content from the X API. Ingestion points: Data is brought into the agent context through commands like xurl read, xurl search, and xurl mentions. Boundary markers: The instructions lack explicit delimiters or 'ignore' directives for content returned by the API. Capability inventory: The skill allows the agent to take actions on the user's behalf, such as posting tweets, deleting posts, and sending DMs. Sanitization: No filtering or sanitization of API responses is described.
- [COMMAND_EXECUTION]: The skill's primary function is to execute the xurl CLI binary with arguments that may be influenced by external input, representing a risk if inputs are not properly validated.
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