Index of every page on this site: llms.txt
All skills
nvidia avatar

/tilegym-cutile-python

@81b3b24
by NVIDIA Corporationnvidia/tilegym847 stars
92
Run it yourself

Expert cuTile programming assistant. Write high-performance GPU kernels using cuTile's tile-based programming model with proper validation and optimization. Supports deep agent orchestration for complex multi-kernel tasks.

  • 37 files
  • 289.3 KB
  • CC-BY-4
  • Updated last month
  • GitHub
Use this Skill: https://skilld.dev/gh/nvidia/tilegym/tilegym-cutile-python

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

BENCHMARK.md

≈1.9k tokens on demand. Your agent reads this file only when SKILL.md points to it.

Skill Benchmark: tilegym-cutile-python

❌ Overall verdict: FAIL — Publication blocked

The skill should be reviewed before publication. Address the blocking findings below, then rerun Skill Evaluator.

Evaluation Metadata

  • Skill: tilegym-cutile-python
  • Evaluation date: 2026-10-16
  • Evaluator version: 1.5.6
  • Agents: Claude Code (aws/anthropic/bedrock-claude-opus-4-8), Codex (openai/openai/gpt-5.5)
  • Tasks: 3 evaluation tasks (2 positive, 1 negative)
  • Dataset digest: sha256:8288ba1fc833c22b44a0957b327f0c48f0c9288983f830241d752d6e9dc68716 (skill-evaluator-dataset-snapshot/1)
  • Attempts per task: 1
  • Environment: k8s-sandbox
  • Tier 2 evidence: required for publication
  • Tier 3 evidence: required for publication

Each task attempt ran in its own isolated sandbox pod.

What This Report Answers

The three-tier evaluation checks whether the skill:

  • is safe to use;
  • produces correct answers;
  • is discovered and activated when needed;
  • helps the agent complete the user's goal and expected workflow; and
  • avoids wasted skill and tool usage.

Results at a Glance

Measure Claude Code (Baseline → Skill Uplift) Codex (Baseline → Skill Uplift)
Overall 91.8% — baseline ran, but no comparable score was available; uplift unavailable 91.9% — baseline ran, but no comparable score was available; uplift unavailable
Security 100.0% → 100.0% (±0.0 points) 100.0% → 100.0% (±0.0 points)
Correctness 100.0% → 100.0% (±0.0 points) 100.0% → 100.0% (±0.0 points)
Discoverability 87.5% — baseline ran, but no comparable score was available; uplift unavailable 88.0% — baseline ran, but no comparable score was available; uplift unavailable
Effectiveness 75.8% → 91.1% (+15.3 points) 84.8% → 84.5% (-0.3 points)
Efficiency 80.3% — baseline ran, but no comparable score was available; uplift unavailable 87.1% — baseline ran, but no comparable score was available; uplift unavailable

How to read this table: baseline is the same task attempted without the target skill. Scores are rounded to one decimal; threshold-adjacent values use additional precision so their displayed band matches the verdict. Uplift is derived from those displayed scores and shown in percentage points.

Example: 47.0% → 92.0% (+45.0 points) means the skill-assisted run scored 92.0%, 45.0 percentage points above its 47.0% no-skill baseline.

A partial dimension was calculated from only the available configured signals; review the detailed report before relying on it.

Token Usage

Actual Tier 3 execution usage is reported for every observed agent/case pair and both conditions.

Agent Dataset case With skill Without skill Delta Change Coverage
claude-code All cases 1,721,825 124,427 +1,597,398 +1283.80% skill 3/3; base 3/3
claude-code 01-write-rmsnorm-cutile-kernel 916,155 35,828 +880,327 +2457.09% skill 1/1; base 1/1
claude-code 02-write-softmax-cutile-kernel 745,855 29,052 +716,803 +2467.31% skill 1/1; base 1/1
claude-code 03-kubernetes-yaml-negative 59,815 59,547 +268 +0.45% skill 1/1; base 1/1
codex All cases 463,706 261,198 +202,508 +77.53% skill 3/3; base 3/3
codex 01-write-rmsnorm-cutile-kernel 168,952 69,425 +99,527 +143.36% skill 1/1; base 1/1
codex 02-write-softmax-cutile-kernel 281,235 178,290 +102,945 +57.74% skill 1/1; base 1/1
codex 03-kubernetes-yaml-negative 13,519 13,483 +36 +0.27% skill 1/1; base 1/1
ALL AGENTS Dataset aggregate 2,185,531 385,625 +1,799,906 +466.75% skill 6/6; base 6/6

Prompt tokens include cached reads, so total tokens are prompt + completion (cached is not added twice). The Efficiency score uses (prompt - cached) + completion. N/A means the relevant trajectory counters were not available; coverage is never estimated.

Tier Status

Tier Purpose Status Evidence
Tier 1 Static validation FAILED 11 validator(s); 28 finding(s)
Tier 2 Semantic deduplication PASSED WITH OBSERVATIONS 2 validator(s); 1 finding(s)
Tier 3 Live agent evaluation PASS 2 agent(s); 3 task(s)

Findings and Observations

<details> <summary>Show detailed findings and successful checks</summary>
  • CRITICAL CONTENT_DEDUP/llm_cluster_count_limit: Tier 2 found more than 50 clusters requiring LLM review. (skills/tilegym-cutile-python)
  • MEDIUM SECURITY/Skill Enumeration (AS3): Agent Snooping: skills/tilegym-cutile-python/SKILL.md (BENCHMARK.md:72)
  • MEDIUM SECURITY/Skill Enumeration (AS3): Agent Snooping: skills/tilegym-cutile-python/SKILL.md (BENCHMARK.md:73)
  • MEDIUM SECURITY/Skill Enumeration (AS3): Agent Snooping: skills/tilegym-cutile-python/SKILL.md (BENCHMARK.md:74)
  • MEDIUM SECURITY/Skill Enumeration (AS3): Agent Snooping: skills/tilegym-cutile-python/SKILL.md (BENCHMARK.md:75)
  • 24 additional finding(s) are available in the full evaluation artifacts.
</details>

Scoring Methodology

<details> <summary>Show dimension definitions, source signals, and thresholds</summary>
Dimension Question Scored signals
Security Is it safe to use? security (100%)
Correctness Is the answer correct? accuracy (100%)
Discoverability Was the right skill loaded when needed? skill_execution (100%)
Effectiveness Did the skill help complete the task? goal_accuracy (50%) + behavior_check (50%)
Efficiency Did it avoid wasted tool calls and token usage? skill_efficiency (50%) + token_efficiency (50%)
  • Dimension bands: PASS at 50% or above; NEUTRAL from 40% to below 50%; FAIL below 40%.
  • Overall Tier 3 lift: PASS at +5 points or more; FAIL at -10 points or less; values between those bands are NEUTRAL.
  • Overall verdict: PASS only when every configured dimension passes for at least one supported agent. Lift is reported as diagnostic evidence and does not override this gate.
  • The 50% attempt pass threshold is a separate per-task gate; it is not the dimension pass threshold.
  • Effectiveness is the equal-weight mean of goal completion (goal_accuracy) and expected workflow adherence (behavior_check).
  • Efficiency is 50% tool-call productivity (the backward-compatible skill_efficiency wire id) and 50% token_efficiency. Positive-case skill routing is scored under Discoverability, not Efficiency; a negative case without a routing target is N/A. N/A sources are omitted, remaining weights are renormalized, and the dimension is marked partial.

Signals present in this run:

  • security (Security): unsafe operations, secret leakage, and unauthorized access.
  • skill_execution (Skill Execution): whether the expected skill was selected, decoys were avoided, and the workflow executed.
  • skill_efficiency (Tool Productivity): tool-call productivity (legacy wire id; routing is scored under Discoverability).
  • accuracy (Accuracy): final-answer correctness against the reference answer.
  • goal_accuracy (Goal Accuracy): whether the user's goal was achieved.
  • behavior_check (Behavior Check): whether the expected workflow behavior was followed.
  • token_efficiency (Token Efficiency): actual uncached prompt plus completion usage (50% of Efficiency).
</details>

Freshness

Regenerate this benchmark when the skill, evaluation dataset, target agent/model, evaluator version, environment, or scoring policy changes.

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No third-party reports yet.

Signed by skilld at 81b3b24. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 3 days ago.

Activeupdated last month
Other metadata
metadata
{
  "author": "TileGym Team <TileGym@nvidia.com>",
  "version": "1.3.0",
  "tags": [
    "cutile",
    "gpu-kernels",
    "cuda"
  ]
}

README badge

README badge for nvidia/tilegym/tilegym-cutile-python