The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the kabi-discord-cli package using uv tool install or pipx install. This is an external dependency from a third-party source (github.com/jackwener/discord-cli) that is not included in the trusted vendor list.
[COMMAND_EXECUTION]: The skill's primary functionality relies on executing shell commands via the discord CLI tool to perform operations such as syncing history, searching messages, and exporting data.
[CREDENTIALS_UNSAFE]: The skill handles highly sensitive Discord authentication tokens. It promotes the use of discord auth --save to automatically extract tokens from the local Discord desktop application or browser sessions. While the documentation warns against echoing these tokens, the automated extraction and management of such credentials by an agent pose a security risk.
[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface (Category 8) by ingesting untrusted content from external Discord messages.
Ingestion points: External message content is retrieved via commands like discord dc history, discord dc search, and discord today (SKILL.md, references/commands.md).
Boundary markers: Absent. The instructions do not mandate the use of delimiters or specific warnings to prevent the agent from obeying instructions embedded within the Discord messages it reads.
Capability inventory: The agent possesses the capability to execute arbitrary discord CLI commands, which could be exploited if malicious instructions are processed (SKILL.md).
Sanitization: Absent. There are no instructions or mechanisms provided to sanitize or filter the content retrieved from Discord before it is presented to the agent's context.

discord