The Agent Skills Directory

[COMMAND_EXECUTION]: The skill allows the execution of arbitrary JavaScript code within the running Obsidian application via the obsidian eval command, which is a standard requirement for live application debugging.
[COMMAND_EXECUTION]: Enables interaction with the Chrome DevTools Protocol through obsidian dev:cdp, allowing for advanced diagnostics such as screenshots and runtime evaluation.
[DATA_EXFILTRATION]: Provides tools to read the application's DOM and internal state, which can expose content from the user's Obsidian vault to the agent during a debugging session.
[PROMPT_INJECTION]: The skill ingests data from console logs, error buffers, and DOM elements, which could contain instructions from untrusted sources like malicious plugins or note content.
Ingestion points: obsidian dev:console, obsidian dev:errors, and obsidian dev:dom commands.
Boundary markers: The instructions do not specify the use of delimiters or warnings to ignore instructions within the ingested data.
Capability inventory: The agent has access to powerful tools including arbitrary JavaScript execution (obsidian eval) and Chrome DevTools Protocol interaction.
Sanitization: No sanitization or validation of external content is defined.

dev-tools