pollo-generate

Pass

Audited by Gen Agent Trust Hub on Aug 5, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to install the pollo CLI via the npm registry (@pollo-ai/cli) or by downloading a shell script from the vendor's official GitHub repository (https://raw.githubusercontent.com/polloaiofficial/cli/main/install.sh).
  • [REMOTE_CODE_EXECUTION]: The standalone installer script is piped directly to the shell for execution (curl ... | sh). This is the vendor's official installation method for the tool required by the skill.
  • [COMMAND_EXECUTION]: The skill drives the pollo CLI tool using the Bash tool to perform account checks, model discovery, and generation tasks. It includes explicit instructions to never read, print, or copy the stored credentials, relying on the CLI's internal authentication handling.
  • [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it interpolates user-supplied text into the --prompt argument of CLI commands.
  • Ingestion points: User-provided text for image and video generation prompts in SKILL.md (Step 3).
  • Boundary markers: Absent; user input is passed directly as a string argument to the command line.
  • Capability inventory: Execution of arbitrary CLI commands via the Bash tool.
  • Sanitization: Absent; the skill instructions do not provide methods for escaping or validating user-provided prompt content.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 5, 2026, 09:36 AM
Security Audit — agent-trust-hub — pollo-generate