ebook-convert
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using the
ebook-convertutility with variable input and output paths (ebook-convert <in> <out>). This pattern creates a command injection surface if the agent processes filenames containing shell metacharacters (e.g., semicolons, backticks, or pipes) without adequate sanitization. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of ebook files (EPUB, MOBI, etc.). Maliciously crafted files could potentially exploit vulnerabilities in the underlying conversion engine or contain instructions designed to influence the agent's behavior.
- Ingestion points: Workflow step 1 and 2 in
SKILL.mdspecify locating and processing external input files. - Boundary markers: No explicit delimiters or warnings to ignore instructions within the ebook content are provided.
- Capability inventory: The skill possesses shell execution capabilities via the
ebook-converttool. - Sanitization: No mention of filename or content sanitization to prevent injection or parser exploitation.
Audit Metadata