# Benchmark Execution

> Benchmark task execution contract: complete every numbered requirement, execute tool calls directly (never plan-only), and verify final state with a direct retrieve/list before responding. Use for every agentic-use benchmark task.

- Skill: `nvidia-nemo/benchmark-execution-2` (Agent Skill)
- Install (CLI): `npx skillmds@latest add nvidia-nemo/benchmark-execution-2`
- Raw SKILL.md: https://api.skillmd.com/api/skills/nvidia-nemo/benchmark-execution-2/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: NVIDIA NeMo (https://skillmd.com/u/nvidia-nemo)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/nvidia-nemo/benchmark-execution-2

---

# Benchmark execution contract

This skill defines the execution requirements that every nemo-studio-assistant run under
`tests/agentic-use/` must satisfy so the canonical gate
(`tests/agentic-use/passrate_token_policy_gate.py`) can score the run on
verifier pass-rate and token totals. See
[`tests/agentic-use/README.md`](../../../../../../tests/agentic-use/README.md)
for the full Run -> Gate -> Optimize loop these tasks plug into.

- Treat `instruction.md` as the task contract: finish all numbered requirements.
- Execute tool calls yourself; do not end with a plan-only response.
- Keep operations minimal and task-focused; avoid unrelated exploration.
- For CRUD-style tasks, if instructions require a final verification resource/state,
  ensure that final state exists before your last response.
- Before final response, run at least one direct verification call that checks the
  required end state from the instruction (for example: retrieve/list/get status).
  Preserve all additional skill-specific verification requirements, including
  requirements for a higher number of verification calls.