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。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/PSPDFKit-labs/agentic-usability --skill export命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
正在显示 SKILL.md
基于 SOC 职业分类
| name | export |
| description | Export a benchmark pipeline as a zip file for sharing or archiving. Excludes cache and large snapshots. |
| argument-hint | [project-directory] [-o output.zip] [-r runId] |
| disable-model-invocation | true |
| allowed-tools | Bash(agentic-usability *) |
Export the pipeline project as a zip archive for sharing or archiving.
echo "Arguments: $ARGUMENTS"
-o, --output <path>: Output zip file path (default: <pipeline-name>-export.zip)-r, --run <runId>: Export only a specific run instead of the entire projectThe zip includes:
config.json — pipeline configurationsuite.json — test suiteresults/ — all run results (judge scores, solutions, logs)cache/** — git repo clones (can be re-fetched)**/*.tar.gz — workspace snapshots (large binary files)<project>/
config.json
suite.json
results/<runId>/
run.json # Run manifest
pipeline-state.json # Pipeline state
report.json # Scorecard
<target>/<testId>/ # Per-test results
Run agentic-usability export -p $ARGUMENTS.
For the full file inventory, see pipeline-guide.md.
Initialize a new agentic-usability benchmark pipeline project. Use when setting up a new SDK benchmark, creating a config.json, or starting a new evaluation project.
Launch an interactive shell inside a microsandbox for debugging. Supports bare mode, executor setup, or judge setup with optional test case scaffolding.
Run the full evaluation pipeline (execute, judge, report) for an SDK usability benchmark. Use when running a complete benchmark end-to-end, resuming an interrupted pipeline, or checking pipeline status.