decodo-web-scraping
Fail
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: Provides an installation command that fetches a shell script from the vendor's GitHub repository and pipes it directly to the shell: 'curl -fsSL https://decodo.github.io/cli/install.sh | sh'.
- [EXTERNAL_DOWNLOADS]: Fetches CLI and MCP tools from the npm registry using 'npm install' and 'npx'. It also references a hosted MCP server at 'https://mcp.decodo.com/mcp'.
- [COMMAND_EXECUTION]: Instructs the agent to execute the 'decodo' command-line interface for scraping, searching, and managing configuration.
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from arbitrary external websites and search engines, which represents a potential attack surface for indirect prompt injection. (1) Ingestion points: Content retrieved via 'decodo scrape' and 'decodo google-search' subcommands. (2) Boundary markers: No specific delimiters or warnings for the agent to ignore instructions within the scraped data are included. (3) Capability inventory: The skill has the ability to execute shell commands (via 'decodo' CLI) and perform network requests (via the API fallback). (4) Sanitization: There are no instructions for sanitizing or escaping the data returned from scraping operations.
Recommendations
- HIGH: Downloads and executes remote code from: https://decodo.github.io/cli/install.sh - DO NOT USE without thorough review
Audit Metadata