Skills
skills/bluewaves-creations/bluewaves-skills/build-skill-zips/Gen Agent Trust Hub

build-skill-zips

Warn

Audited by Gen Agent Trust Hub on Mar 28, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute a local script located at scripts/build-skill-zips.sh.
  • [PROMPT_INJECTION]: The skill interpolates user-supplied data ($ARGUMENTS) directly into a shell command line (bash scripts/build-skill-zips.sh $ARGUMENTS). This surface is vulnerable to indirect prompt injection and command injection.
  • Ingestion points: User input is ingested via the $ARGUMENTS variable in SKILL.md.
  • Boundary markers: No delimiters or instructions to ignore embedded commands are present around the interpolated arguments.
  • Capability inventory: The skill has access to the Bash tool, allowing for arbitrary command execution on the host system.
  • Sanitization: There are no instructions provided to the agent to sanitize, validate, or escape the content of $ARGUMENTS before shell execution.
  • [REMOTE_CODE_EXECUTION]: The skill relies on an external script file (scripts/build-skill-zips.sh) that is not included in the provided context, making its behavior and safety unverifiable during this audit.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 28, 2026, 10:15 PM