my-cubrid-skills-create
Pass
Audited by Gen Agent Trust Hub on Apr 8, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) as it automatically extracts and processes instructions from conversation history and user arguments to generate new executable skill code.
- Ingestion points: The skill reads the
$ARGUMENTSvariable and the current conversation history (tools used, commands, steps) to draft the new skill's content. - Boundary markers: No explicit delimiters or instructions are provided to the agent to isolate or ignore potentially malicious instructions embedded within the extracted workflow data.
- Capability inventory: The skill can create new directories and execute a specialized creation tool (
/skill-creator) to write persistent files to/home/vimkim/temp/my-cubrid-skills/. - Sanitization: The skill lacks logic to validate or sanitize the data extracted from the conversation before using it to generate new SKILL.md and script files.
- [COMMAND_EXECUTION]: The skill performs directory creation on the local file system and provides instructions for the user to execute local build/install scripts (e.g.,
just install). These operations are consistent with its stated purpose of local development and repository management.
Audit Metadata