lunarcrush

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFEPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [DATA_EXFILTRATION]: The skill performs network requests to https://lunarcrush.com/api4 to retrieve social sentiment data. This is the primary function of the skill; however, the domain is not among the whitelisted developer service domains.
  • [PROMPT_INJECTION]: The skill processes untrusted content from external social media platforms, creating a surface for indirect prompt injection where malicious instructions embedded in social posts could influence the agent.
  • Ingestion points: Untrusted social media posts and news summaries are retrieved in tools/topics.py and tools/creators.py via tools such as lunar_topic_posts and lunar_creator_posts.
  • Boundary markers: The skill does not implement explicit boundary markers or instructions to the agent to disregard commands within the fetched data.
  • Capability inventory: Analysis of lunarcrush.py and the tools/ directory indicates the skill does not have dangerous capabilities like file writing, command execution, or dynamic code evaluation.
  • Sanitization: No sanitization or filtering of the retrieved social content is performed before returning it to the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 10:07 PM