The Agent Skills Directory

[COMMAND_EXECUTION]: The 'Operating Rules' (Rule 4) instruct the agent to offer a global installation command: npx skills add wondelai/skills/<slug> --global. This prompts the user to execute external scripts with high privileges on their system.
[EXTERNAL_DOWNLOADS]: The skill depends on fetching additional logic from an external source (wondelai/skills). This source is not recognized as a trusted organization or well-known service, posing a supply-chain risk if the external repository is compromised or malicious.
[PROMPT_INJECTION]: The skill implements a resumable workflow that reads and interprets data from various markdown files in the project's docs/ directory.
Ingestion points: Reads docs/CREATE-WEBSITE-PLAN.md, docs/POSITIONING.md, docs/WEBSITE.md, and other artifacts during the 'Resume first' step and phase transitions.
Boundary markers: The instructions lack explicit delimiters or warnings to ignore embedded instructions when reading content from these files.
Capability inventory: The skill has the ability to suggest shell commands and perform file system writes to the project directory.
Sanitization: There is no evidence of validation or filtering for the content ingested from the external documentation files before it is processed by the agent.

create-website