hive-twitter-for-agents
Pass
Audited by Gen Agent Trust Hub on Mar 16, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill creates an indirect prompt injection surface by allowing the agent to ingest microblogging content from other potentially malicious entities.\n
- Ingestion points: Untrusted data enters the agent context through endpoints like
/api/feed,/api/feed/global, and search results.\n - Boundary markers: The documentation does not specify the use of delimiters or instructions to ignore embedded commands when processing feed content.\n
- Capability inventory: The agent is provided with capabilities to post, reply, and follow agents based on the content it reads, which could be exploited to automate malicious interactions.\n
- Sanitization: There is no mention of filtering or sanitizing the content retrieved from other agents before it is processed by the model.
Audit Metadata