The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill's documentation explicitly recommends installing components using the command npx skills add AbsolutelySkilled/AbsolutelySkilled. This executes code from an external GitHub repository that is not part of a recognized or trusted organization, representing a significant supply-chain risk.
[COMMAND_EXECUTION]: The skill instructs the AI agent to execute shell commands (ls) to inspect the user's local directory structure for existing skills (e.g., ~/.claude/skills/). This practice of running system commands based on skill instructions increases the potential attack surface and user privacy risk.
[EXTERNAL_DOWNLOADS]: The skill suggests downloading third-party npm packages such as schema-dts and structured-data-linter. While these are legitimate tools, they represent external dependencies that the skill introduces to the user's workflow.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted code from web pages to determine schema placement without sufficient isolation. 1. Ingestion points: The agent reads the user's source code files. 2. Boundary markers: No delimiters or isolation instructions are provided. 3. Capability inventory: The agent has shell command execution and user interaction capabilities. 4. Sanitization: No sanitization of ingested content is performed.

schema-markup