Explore an unfamiliar area of the Stave codebase quickly. Use when the request asks for workspace structure, architecture overview, relevant files, entrypoints, "where should I look first", "관련 코드 확인", "구조 파악", or broad code search before implementation.
89
84%
Does it follow best practices?
Impact
99%
1.05xAverage score across 3 eval scenarios
Passed
No known issues
IPC layer navigation report
AGENTS.md or arch docs first
100%
100%
architecture/index.md consulted
100%
100%
architecture/entrypoints.md consulted
100%
100%
IPC subsystem named
100%
100%
window-api.d.ts identified
100%
100%
electron/main/ipc/schemas.ts identified
100%
100%
Contract boundaries described
100%
100%
Follow-up files or queries provided
100%
100%
No full app.store.ts load
100%
100%
Scope kept narrow
100%
87%
Explore subagent or targeted approach
100%
100%
Provider event normalization flow trace
Arch docs before impl files
100%
100%
provider.types.ts identified
100%
100%
schemas.ts identified with types
100%
100%
provider-event-replay.ts identified
100%
100%
Both SDK runtimes examined together
100%
100%
Subsystem named
100%
100%
Contract boundaries described
100%
100%
Follow-up files or queries
100%
100%
No broad app.store.ts load
75%
100%
Targeted vs broad search
100%
100%
Scope kept to provider pipeline
100%
83%
Stave auto feature architecture map
Arch docs read first
80%
100%
architecture/index.md consulted
100%
100%
architecture/entrypoints.md consulted
100%
100%
Agent/Explore subagent used
0%
100%
No rg/Glob chain as primary strategy
100%
100%
Subsystem named
100%
100%
Key files listed
100%
100%
Contract boundaries described
100%
100%
Follow-up files or searches
100%
100%
Output scoped to feature
100%
100%
No whole-file bulk loads
100%
100%
e275ad5
Table of Contents
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.