The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill instructs the agent to recommend a shell command (npx skills add wondelai/skills/<slug> --global) that downloads and executes code from an untrusted external source. The use of npx to fetch and install global packages from an unverified repository constitutes a significant remote code execution risk.
[COMMAND_EXECUTION]: The skill provides specific shell commands for the user or agent to run, including npx commands with global installation flags, which could alter the system environment.
[EXTERNAL_DOWNLOADS]: The skill references and encourages the download of external skill packages from an unverified source (wondelai) not recognized as a trusted vendor.
[PROMPT_INJECTION]: The skill's architecture is vulnerable to indirect prompt injection.
Ingestion points: It is designed to ingest and process external content such as website URLs, live copy, CSS, and visitor input during the intake and analysis phases.
Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from following instructions embedded within the ingested data.
Capability inventory: The skill has the capability to write to the local filesystem (docs/ directory) and recommends shell command execution.
Sanitization: There are no documented sanitization or validation procedures for the external assets processed by the skill.

improve-website