morning-ai
Warn
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill's installation process requires cloning an external repository (
github.com/octo-patch/MorningAI) to access the scripts used for data collection and report generation. - [INDIRECT_PROMPT_INJECTION]: The skill aggregates and summarizes content from six external platforms (Reddit, Hacker News, GitHub, HuggingFace, arXiv, and X/Twitter), which provides a vector for indirect prompt injection.
- Ingestion points: Data is collected from 80+ entities across multiple social and technical news sources.
- Boundary markers: There are no instructions or delimiters specified to prevent the agent from interpreting instructions found within the collected external data.
- Capability inventory: The skill uses external Python scripts to collect, score, and format news items into reports.
- Sanitization: No sanitization or filtering logic is mentioned to mitigate the risk of malicious instructions being processed from the untrusted data sources.
Audit Metadata