The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to use npx @hackwiki/cli, which fetches the package from the npm registry at runtime.
[REMOTE_CODE_EXECUTION]: Running npx @hackwiki/cli involves executing code downloaded from an external source. While this is the intended purpose of the skill, it remains an unverified external dependency.
[COMMAND_EXECUTION]: The skill utilizes shell commands such as hackwiki session, hackwiki search, and hackwiki page to manage wiki content.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection by reading external data.
Ingestion points: Data is ingested through the hackwiki page read <noteId> command, which fetches Markdown content from an external HackMD instance.
Boundary markers: There are no instructions or delimiters defined to help the agent distinguish between the skill's instructions and potentially malicious instructions embedded within the fetched notes.
Capability inventory: The agent has the capability to write, update, and rename pages, as well as execute CLI commands.
Sanitization: The instructions do not specify any sanitization or validation of the content retrieved from the notes before it is processed by the agent.

hackwiki