blogwatcher
Warn
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill metadata triggers the installation of an external Go module from an unverified GitHub repository (github.com/Hyaxia/blogwatcher). This binary handles network communications and feed parsing and is not from a recognized vendor.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external RSS and Atom feeds, which are untrusted sources. 1. Ingestion points: The scan and articles commands fetch and display content from arbitrary remote URLs. 2. Boundary markers: No delimiters or instructions are provided to help the agent distinguish between blog content and system instructions. 3. Capability inventory: The skill uses the blogwatcher binary for network access and local state management. 4. Sanitization: There is no evidence of validation or sanitization of the content retrieved from the feeds.
- [COMMAND_EXECUTION]: The skill requires execution of the blogwatcher CLI for all operational tasks, involving network requests and file system interactions.
Audit Metadata