The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources and existing notes, creating a potential attack surface for indirect instructions to influence agent behavior.
Ingestion points: Untrusted data enters the agent context via the import_web_page tool (which fetches external URL content) and the read_note / read_blocks tools (which read stored note content).
Boundary markers: The instructions do not provide explicit delimiters or warnings to the agent to ignore potentially malicious instructions embedded within the processed content.
Capability inventory: The skill provides significant capabilities, including edit_block, batch_edit, delete_note, and manage_note_tags, which could be abused if the agent is influenced by injected content.
Sanitization: There is no mention of content sanitization or validation of the fetched/read data before it is presented to the agent.
[COMMAND_EXECUTION]: The skill's documentation (specifically CLI_REFERENCE.md) describes the installation and use of a command-line interface wpsnote-cli. It provides specific paths and commands for Windows, macOS, and Linux environments, which are used to interact with the WPS Note service.

wps-note