The Agent Skills Directory

[Indirect Prompt Injection] (LOW): The skill ingests untrusted external data via command arguments which could influence AI output.
Ingestion points: skill.json and docs/QUICKSTART.md show commands accepting external files like data sales.csv and knowledge faq.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the provided skill definitions.
Capability inventory: The skill is capable of generating functional HTML, JavaScript, and document content.
Sanitization: No sanitization or validation of the input file content is defined in the available files.
[Remote Code Execution] (SAFE): No instances of downloading and piping remote scripts to shells or interpreters were found.
[Credentials Unsafe] (SAFE): No hardcoded secrets, API keys, or private configuration files were detected in the repository structure.

easy-app-maker