The Agent Skills Directory

[DYNAMIC_EXECUTION]: The provided templates for Discord.js and Pycord use dynamic file loading to register commands and extensions.
In references/detailed-guide.md, the JavaScript foundation uses require(filePath) where filePath is dynamically computed by iterating over the src/commands/ and src/events/ directories.
In references/detailed-guide.md, the Python foundation uses bot.load_extension() with paths derived from os.listdir("./cogs").
[INDIRECT_PROMPT_INJECTION]: The skill guides the creation of bots that ingest untrusted data from Discord users (messages, interactions), creating a potential attack surface.
Ingestion points: External data enters the agent context through Discord Gateway events such as interactionCreate (JavaScript) and slash command callbacks (Python) found in references/detailed-guide.md.
Boundary markers: Explicit delimiters or 'ignore' instructions are absent in the boilerplate code, leaving boundary enforcement to the developer's implementation.
Capability inventory: The templates include capabilities to respond to interactions, send messages, and access guild metadata via requested intents.
Sanitization: No input sanitization or validation logic is included in the basic structural templates.

discord-bot-architect