The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill guides the user to clone a repository from an external GitHub account (https://github.com/mengjian-github/openclaw101.git) that is not associated with the skill author or a trusted organization.
[COMMAND_EXECUTION]: The instructions include running npm install and npm run dev. Running npm install on an untrusted repository can lead to the execution of malicious scripts defined in the package.json file (e.g., preinstall or postinstall hooks).
[INDIRECT_PROMPT_INJECTION]: The skill defines a workflow for aggregating external web resources (tutorials, guides) into a local database file (src/data/resources.ts).
Ingestion points: The src/data/resources.ts file serves as the ingestion point for untrusted external data (titles, descriptions, URLs).
Boundary markers: There are no boundary markers or instructions to the agent to disregard malicious directives embedded in the resource descriptions.
Capability inventory: The environment allows shell command execution (npm, git) and file system modifications.
Sanitization: The skill provides no mechanisms for sanitizing or validating the content of the external resources before they are processed by the agent or the platform.

openclaw101-resource-platform