ship-learn-next
Pass
Audited by Gen Agent Trust Hub on Mar 1, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits vulnerability to indirect prompt injection as it is designed to ingest and process untrusted external data (transcripts, articles, tutorials) provided by users.
- Ingestion points: The
Readtool is used to access content from a user-specifiedFILE_PATH. - Boundary markers: There are no instructions to use delimiters or ignore potential commands within the external content during the analysis phase.
- Capability inventory: The skill possesses
ReadandWritetool permissions, allowing it to access local files and create new ones. - Sanitization: No validation or sanitization of the input file's content is implemented before processing by the LLM.
- [COMMAND_EXECUTION]: The skill requests a file path from the user and uses a shell-like representation (
FILE_PATH="/path/to/content.txt") to define input. While the tools are restricted toReadandWrite, an agent following these instructions might be manipulated into reading sensitive system files (e.g., configuration or credential files) if a malicious path is provided.
Audit Metadata