blogwatcher
Warn
Audited by Gen Agent Trust Hub on Mar 11, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the installation of a Go binary from a third-party GitHub repository not associated with trusted organizations.
- Evidence: The installation instructions specify
github.com/Hyaxia/blogwatcher/cmd/blogwatcher@latestas the source for the tool. - [COMMAND_EXECUTION]: The skill's operation depends on the execution of the third-party
blogwatcherCLI tool. - Evidence: Multiple commands such as
blogwatcher scanandblogwatcher articlesare documented for use. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its processing of untrusted external content.
- Ingestion points: Data is ingested from external RSS/Atom feeds via the
scanandarticlessubcommands. - Boundary markers: Absent. The skill does not provide any instructions or delimiters to help the agent distinguish between data and potential commands within the feed content.
- Capability inventory: The skill utilizes the
blogwatcherbinary to fetch and display content from external URLs. - Sanitization: Absent. The skill definition includes no logic for sanitizing or validating the content retrieved from external feeds.
Audit Metadata