bx-rss
Pass
Audited by Gen Agent Trust Hub on Apr 15, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill facilitates the ingestion of untrusted data from external RSS/Atom feeds through the
rss()function andbx:feedcomponent. Ingestion points: External data enters the agent context through remote feed URLs parsed inSKILL.md. Boundary markers: No explicit delimiters or instructions are provided to the agent to isolate or ignore embedded instructions within the parsed feed content. Capability inventory: The skill includes capabilities for network requests (bx:feed) and file system operations (fileWrite). Sanitization: No sanitization of ingested feed data is demonstrated before the data is processed or output. - [EXTERNAL_DOWNLOADS]: Fetches external resources. The skill instructs the user to install the
bx-rssmodule from the vendor's registry and performs network requests to retrieve feed content from remote URLs as part of its primary functionality. - [DATA_EXFILTRATION]: Network Operations. The skill performs network GET requests to external domains to retrieve RSS and Atom feeds.
- [COMMAND_EXECUTION]: Installation Commands. The skill provides shell commands for installing the
bx-rssmodule via theinstall-bx-moduletool andbox(CommandBox) package manager.
Audit Metadata