一键导入
zihuan-test
Run zihuan-next tests. Use this skill when asked to run tests, debug test failures, or find and execute specific test cases in the project.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Run zihuan-next tests. Use this skill when asked to run tests, debug test failures, or find and execute specific test cases in the project.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Develop complete agents in zihuan-next. Use this skill when asked to create, configure, or modify agents — including Brain setup, LLM binding, built-in tool loading, node graph tool definitions, and agent lifecycle management.
Develop tools and agent capabilities for the zihuan-next agent system. Use this skill when asked to create new tools for the QQ chat agent, implement FunctionTool traits, define embedded subgraph tools, or work with the Brain tool-calling loop.
Build the zihuan-next project from source. Use this skill when asked to build, compile, or run the Rust project, including CUDA/Metal feature builds, frontend builds, and Docker builds.
Develop the zihuan-next WebUI. Use this skill when asked to modify the admin panel (Vue 3), the visual node-graph editor (Litegraph.js), add UI features, or change the TypeScript frontend code in webui/.
Lint and format zihuan-next code. Use this skill when asked to check for warnings, run clippy, format code, or fix lint issues.
Develop new node types for the zihuan-next DAG graph engine. Use this skill when asked to create, modify, or debug graph nodes — including port definitions, node registration, Node trait implementation, and lifecycle management.
| name | zihuan-test |
| description | Run zihuan-next tests. Use this skill when asked to run tests, debug test failures, or find and execute specific test cases in the project. |
The project uses Rust's built-in test framework (#[test]). Tests are organized as:
#[cfg(test)] mod tests)tests/ directories at crate roots# Run all tests across the entire workspace
cargo test
# Run tests for a specific crate
cargo test -p zihuan_service
cargo test -p node_macros
cargo test -p zihuan_agent
cargo test -p ims_bot_adapter
# Run a specific test by name
cargo test -p zihuan_service test_name
# Run tests with stdout/stderr output visible
cargo test -- --nocapture
# Run tests, showing output and not capturing test results
cargo test -- --show-output
# Run ignored tests
cargo test -- --ignored
| Crate | Integration tests |
|---|---|
zihuan_service | zihuan_service/tests/ — QQ agent tool inputs, image understanding, deep search workflow |
node_macros | node_macros/tests/ — flow and node macros |
ims_bot_adapter | ims_bot_adapter/tests/ — adapter and message tests |
#[test]test_<what>_<expected_behavior> (e.g., test_tool_input_expands_reply_source_images_into_top_level_message_list)tests/ directories at the crate root#[cfg(test)] mod tests { ... } within source files#[test]
fn tool_input_expands_reply_source_images_into_top_level_message_list() {
let event = build_reply_image_event();
let expanded = expand_message_event_for_tool_input(&event);
assert!(/* ... */);
}
cargo test with -p <crate> will resolve correctly from there.-p <crate> for fast feedback — avoids compiling and testing unrelated crates.-- --nocapture when debugging to see println! / dbg! output.cargo test > test_output.txt 2>&1.zihuan_service/tests/ test QQ chat agent tool inputs and workflow definitions — these are the most domain-relevant tests.node_macros/tests/ validate the node_input! / node_output! proc macros — run these when changing port definitions or the macro crate.