Tauri 2 all-in-one developer hub: task board backed by an Obsidian vault and Claude Code plugin marketplace
Agent
Broad, read-only code investigation and reference tracing. Use when answering a question means sweeping many files, directories, or naming conventions and you only need the conclusion — not to edit code. Returns findings as filepath:linenumber references.
Agent
Implement large or uncertain changes that span multiple files or require debugging and trial-and-error. Use when the work is substantial enough that an isolated context with its own iteration loop pays off.
Agent
Implement changes whose specification is already settled. Use for mechanical or well-scoped edits where the approach is decided and little trial-and-error is expected. Can batch several small related tasks in one run.
Agent
Run tests (or builds/linters) and report a concise pass/fail summary. Use to execute a known command and fold only the outcome back into the main session, keeping verbose output out of the main context.
Agent
A narrowly scoped editing persona for small changes to one or two existing files. It requires reading files first, applying a minimal edit, and checking the result afterward.
Agent
A read-only coding assistant profile that locates definitions, callers, references, tests, and imports, then reports their file and line locations in a compact format.
Agent
Derived from genshijin (MIT, InterfaceX-co-jp) — https://github.com/InterfaceX-co-jp/genshijin.