zsteg
Pass
Audited by Gen Agent Trust Hub on Jun 19, 2026
Risk Level: SAFENO_CODECOMMAND_EXECUTION
Full Analysis
- [NO_CODE]: The skill consists entirely of a markdown file providing documentation and usage examples. No scripts, binaries, or configuration files are included in the skill package.
- [COMMAND_EXECUTION]: The instructions describe running shell commands like
zsteg,file, andstrings. These are standard command-line tools for security analysis and are presented as manual workflows for the agent to follow. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of image files. This represents an attack surface where an adversary could embed malicious instructions in an image's bit-planes to influence the agent when it reads the tool's output.
- Ingestion points: Analysis of PNG/BMP files via
zsteg(SKILL.md). - Boundary markers: Absent. The instructions do not specify using delimiters or provide warnings for the agent to ignore instructions found within analyzed image data.
- Capability inventory: The agent is instructed to use
zsteg,file,strings,foremost,steghide, andstegseek(SKILL.md). - Sanitization: No sanitization or validation of the extracted data is implemented.
Audit Metadata