prd-generator
Fail
Audited by Gen Agent Trust Hub on Feb 15, 2026
Risk Level: HIGHCOMMAND_EXECUTION
Full Analysis
- COMMAND_EXECUTION (HIGH): Vulnerable command injection point in
scripts/generate_prd.shwithin theprompt_inputfunction. - Evidence: The script uses
eval "$var_name='$input'"to assign user-provided input to variables. An attacker providing input containing a single quote and a semicolon (e.g.,'; touch /tmp/pwned; ') can execute arbitrary shell commands in the context of the user running the script. - Context: The
evalstatement is executed for every piece of information gathered during the interactive PRD generation process. - COMMAND_EXECUTION (MEDIUM): Local file overwrite vulnerability in
scripts/generate_prd.sh. - Evidence: The script allows the user to specify an
OUTPUT_FILEpath which is subsequently used in acat > "$OUTPUT_FILE"operation. - Context: This allows an attacker to overwrite sensitive local files (like
~/.bashrcor~/.ssh/authorized_keys) if the script is run in an environment where the agent or user has write permissions to those locations.
Recommendations
- AI detected serious security threats
Audit Metadata