gen-ai-use
Fail
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONDATA_EXFILTRATIONPERSISTENCEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill recommends installing the Picsart CLI tool via a shell script hosted on the author's official domain (https://picsart.com/gen-ai-cli/install.sh) which is piped directly to the bash shell for execution.
- [DATA_EXFILTRATION]: The CLI tool manages sensitive authentication tokens by reading and writing to a dedicated configuration file at ~/.gen-ai/credentials.json in the user's home directory.
- [PERSISTENCE]: The skill's documentation provides specific instructions for users to add execution commands to their shell profile files (~/.bashrc and ~/.zshrc) to enable persistent tab-completion functionality for the CLI.
- [DYNAMIC_EXECUTION]: The batch generation documentation includes command-line snippets that use node -e to dynamically execute JavaScript code for processing and summarizing JSON result files.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input in the form of natural language prompts and media files provided via CLI arguments, creating a surface for indirect prompt injection.
- Ingestion points: Prompt strings provided to the --prompt flag and file paths for media provided to flags like --image, --video, and --audio.
- Boundary markers: No explicit delimiters or boundary warnings are provided in the skill's instructions.
- Capability inventory: The skill is configured to use the terminal to execute CLI commands and manage local files via Bash, Grep, and Glob tools.
- Sanitization: The skill does not implement custom sanitization, relying instead on the security guardrails of the underlying Picsart CLI and API.
Recommendations
- HIGH: Downloads and executes remote code from: https://picsart.com/gen-ai-cli/install.sh - DO NOT USE without thorough review
Audit Metadata