The Agent Skills Directory

[COMMAND_EXECUTION]: The skill directs the agent to use the Bash tool to execute a directory search (find /path/to/project -maxdepth 3 -name .obsidian -type d). This command is used to locate the Obsidian configuration folder to ensure that file paths in the generated canvas are relative to the vault root. While functionally necessary, it involves executing shell commands in the local environment.
[EXTERNAL_DOWNLOADS]: During Phase 4 (Lesson Generation), the skill utilizes WebSearch and WebFetch to retrieve technical facts, formulas, and standards from external websites. This is intended to ensure pedagogical accuracy but involves ingesting content from untrusted external sources.
[INDIRECT_PROMPT_INJECTION]: The use of web-fetched data creates a surface for indirect prompt injection.
Ingestion points: Technical data is retrieved from the web via WebFetch and incorporated into lesson markdown files.
Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions to wrap the external content.
Capability inventory: The skill has access to Write for file creation and Bash for command execution.
Sanitization: The instructions lack requirements for sanitizing or filtering external content before it is processed by the agent.

create-learning-map