rss-digest
Warn
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- EXTERNAL_DOWNLOADS (MEDIUM): The skill instructions specify the installation of the
feedCLI tool fromodysseus0/tap/feedvia Homebrew orgithub.com/odysseus0/feedvia Go. These sources are not within the trusted repository list, making the dependency unverifiable. - COMMAND_EXECUTION (LOW): The skill relies on executing local CLI commands through the
feedbinary to manage and read RSS feeds. While functional, it assumes the integrity of the downloaded third-party executable. - PROMPT_INJECTION (MEDIUM): The skill is susceptible to Indirect Prompt Injection (Category 8) because it fetches and processes untrusted data from external RSS feeds.
- Ingestion points: Workflow step 4 fetches full content from external URLs or via the
feed get entry <id>command. - Boundary markers: No markers or delimiters are used to separate external RSS content from the agent's instructions.
- Capability inventory: The agent can execute CLI commands and perform web fetches, providing a pathway for malicious instructions embedded in RSS feeds to influence agent actions.
- Sanitization: There is no evidence of sanitization or filtering applied to the external content before summarization.
Audit Metadata