writing-core
Warn
Audited by Gen Agent Trust Hub on Jul 20, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell and PowerShell scripts located in the 'research-writing-skill/scripts/' directory. Running external scripts on the host system is a high-privilege activity that can lead to arbitrary code execution if the scripts are modified or if they handle input unsafely.
- [COMMAND_EXECUTION]: The PowerShell execution command specifically includes the '-ExecutionPolicy Bypass' flag. This is a privilege escalation pattern that intentionally disables PowerShell's security configuration, allowing the execution of unsigned or restricted scripts that would otherwise be blocked by system policy.
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection (Category 8) because it processes untrusted user-supplied academic papers and manuscripts.
- Ingestion points: Academic paper content and file paths provided by the user (identified in SKILL.md Section 7).
- Boundary markers: None. The skill does not define delimiters or instructions to ignore embedded commands within the papers being analyzed.
- Capability inventory: The skill has the ability to execute shell scripts (bash), PowerShell scripts, and read/write to the file system (SKILL.md Section 7 and 8).
- Sanitization: None. There is no evidence of input validation or sanitization before passing filenames or content to the style-checking scripts.
Audit Metadata