skills/agentskillexchange/skills/Monitor cross-platform trends and alert agents with TrendRadar/Gen Agent Trust Hub
Monitor cross-platform trends and alert agents with TrendRadar
Warn
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: MEDIUMMETADATA_POISONINGINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [METADATA_POISONING]: The skill's YAML frontmatter includes deceptive metrics, claiming 59,359 GitHub stars and a 'security_reviewed' verification status. These claims are not supported by the linked repository and appear to be fabricated to falsely enhance the perceived popularity and safety of the skill.
- [INDIRECT_PROMPT_INJECTION]: The skill monitors multi-platform feeds and summarizes signals with AI, creating a vulnerability to indirect prompt injection. Malicious instructions embedded in external feed data could influence the agent's summaries or cause it to execute unintended actions through the MCP interface.
- Ingestion points: Multi-platform external feeds.
- Boundary markers: None identified in the provided instructions.
- Capability inventory: Summarization, alerting, and trend analysis exposure via MCP.
- Sanitization: No evidence of input sanitization or filtering logic.
- [EXTERNAL_DOWNLOADS]: The installation process involves cloning code from GitHub (
sansan0/TrendRadar) and pulling images from Docker Hub (wantcat/trendradar). The discrepancy between the GitHub author name and the Docker namespace is a common supply chain risk factor that requires manual verification of the container contents.
Audit Metadata