sjinn-image-generation
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data which could potentially contain malicious instructions aimed at the agent.
- Ingestion points: The
SKILL.mdfile defines a workflow that accepts user-provided URLs and local file paths via the--imageflag in thesjinn image generatecommand. - Boundary markers: The instructions do not define delimiters or specific warnings to ignore instructions that might be embedded in image metadata or associated data streams.
- Capability inventory: The skill uses the
sjinnCLI to execute subprocesses, access the local filesystem for reading (--image) and writing (--download), and perform network operations. - Sanitization: There is no explicit requirement for the agent to sanitize or validate the content of the provided image URLs or files before they are passed to the CLI tool.
- [COMMAND_EXECUTION]: The skill relies on constructing shell commands using user-supplied input.
- The instructions direct the agent to build commands such as
sjinn image generate --prompt "<user_text>". If the input is not properly escaped by the agent platform, it creates a surface for command injection vulnerabilities where a user could potentially execute arbitrary shell commands.
Audit Metadata