aggregating-crypto-news
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- PROMPT_INJECTION (HIGH): High risk of Indirect Prompt Injection (Category 8). The skill possesses an ingestion surface for untrusted data from 50+ external RSS feeds without specified boundary markers or sanitization. Combined with its capability to execute shell commands and write to the local filesystem (via the --output parameter), a compromised or malicious feed could override agent instructions to perform unauthorized actions.
- EXTERNAL_DOWNLOADS (MEDIUM): The skill requires the installation of external Python libraries
feedparserandrequestsfrom public registries (PyPI). (Category 4). These are not within the defined trusted source scope and should be verified for integrity. - COMMAND_EXECUTION (LOW): The skill uses the Bash tool to execute a local Python script with arguments derived from user input. (Category 5). While typical for functionality, it increases the attack surface if parameters are not strictly validated.
Recommendations
- AI detected serious security threats
Audit Metadata