aaj-marketing-psychology
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions direct the agent to execute a local Node.js script (
pattern-check.js) using shell commands. This script is used to perform technical analysis on marketing copy provided by the user. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted user data, such as landing page copy and emails, which could contain malicious instructions. * Ingestion points: User-provided copy is passed to the analysis script via CLI arguments or local file paths. * Boundary markers: There are no explicit delimiters or instructions to the agent to disregard embedded commands in the analyzed text. * Capability inventory: The skill utilizes shell execution and local file system access (
fs.readFileSync). * Sanitization: The included script does not sanitize for prompt injection, only for specific marketing patterns. - [DYNAMIC_EXECUTION]: The skill executes an included Node.js script at runtime to process strings and files. The script uses standard Node.js APIs to perform regex-based pattern matching on user input.
Audit Metadata