The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The install-skill command in instructions/install-skill.md utilizes npx skills add to fetch and install the skill from the author's npm/GitHub repository (baobao700508/unofficial-remnote-bridge-cli).
[REMOTE_CODE_EXECUTION]: The skill manages 'addons' (extensions) through the addon command (instructions/addon.md), such as remnote-rag, which are separate components installed at runtime. Furthermore, the connect --dev mode automatically installs Node.js dependencies for the plugin.
[COMMAND_EXECUTION]: The skill performs various shell operations to manage the lifecycle of the bridge, including starting/stopping a background daemon, executing npx for installation, and using system commands like kill or taskkill during the clean process to terminate residual processes.
[DATA_EXPOSURE]: The setup command (instructions/setup.md) manages a local Chrome profile in ~/.remnote-bridge/chrome-profile to store RemNote authentication sessions and cookies, which are then reused by the headless connection mode.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the user's RemNote knowledge base via read-rem, read-tree, and search operations. This data is processed in the agent's context and can influence future actions. The skill lacks explicit boundary markers or sanitization instructions for this external content, while possessing powerful capabilities like edit-rem, edit-tree, and clean which could be abused if malicious instructions are present in the notes.

remnote-bridge