메뉴
Open the web UI to visually inspect, edit, and run the benchmark pipeline. Use when the user wants a visual interface for their pipeline.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
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.
Execute benchmark test cases in sandboxed environments with AI agents. Spins up microsandbox containers for each test case and extracts solutions.
Export a benchmark pipeline as a zip file for sharing or archiving. Excludes cache and large snapshots.
Generate SDK usability test cases by exploring source code. Use when creating benchmark test suites, generating test cases for an SDK, or when the user wants to create evaluation scenarios.
| name | inspect |
| description | Open the web UI to visually inspect, edit, and run the benchmark pipeline. Use when the user wants a visual interface for their pipeline. |
| argument-hint | [project-directory] [--port 7373] |
| disable-model-invocation | true |
| allowed-tools | Bash(agentic-usability *) Read Glob |
Launch the web-based inspector for the benchmark pipeline.
echo "Arguments: $ARGUMENTS"
--port <number>: Port for the local server (default: 7373)The web UI serves data from the project directory:
<project>/
config.json # Pipeline configuration
suite.json # Test suite (array of test cases)
results/
<runId>/ # e.g. run-2026-04-25T10-30-00-000Z
run.json # Run manifest (id, targets, testCount, label)
pipeline-state.json # Pipeline progress tracker
report.json # Aggregate scorecard (if pipeline completed)
<target>/<testId>/ # Per-test results
generated-solution.json # Agent's solution
judge.json # Judge scores
agent-notes.md # Agent's working notes
agent-output.log # Raw output
agent-session.jsonl # Agent conversation log
judge-session.jsonl # Judge conversation log
results/ with a run.json are runspipeline-state.json to see if a run is complete (stage: "report") or pausedRun agentic-usability inspect -p $ARGUMENTS to start the server. It opens the browser automatically. Press Ctrl+C to stop.
For the full file inventory, see pipeline-guide.md.