Pass
Audited by Gen Agent Trust Hub on Feb 25, 2026
Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: Utilizes the
curlcommand-line tool to interact with the 6551 API endpoints for data retrieval. - [DATA_EXFILTRATION]: Conducts network operations to
ai.6551.ioto transmit authentication tokens and fetch data. This behavior is the primary intended function of the skill and aligns with its stated purpose of providing Twitter data access. - [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection by ingesting untrusted data from Twitter, such as tweet text and user profile descriptions.
- Ingestion points: Data returned from API endpoints at
https://ai.6551.io/open/*. - Boundary markers: None identified in the skill instructions or code examples.
- Capability inventory: The skill only performs data retrieval and does not possess capabilities for file system modification, arbitrary command execution on fetched data, or secondary network exfiltration.
- Sanitization: No explicit sanitization or filtering of retrieved social media content is performed within the skill scripts.
Audit Metadata