ci: trigger numerical optimization skill evals

[ramakrishnap-nv]

Trivial whitespace change to trigger CI evals for the numerical optimization formulation skill.

[ramakrishnap-nv]

/nvskills-ci

[coderabbitai]

📝 Walkthrough

Walkthrough

The cuopt-numerical-optimization-formulation skill documentation was updated to reflect a new evaluation run. The Use Case description was rewritten to emphasize parsing unstructured problems into LP/MILP/QP formulations. Skill Output types were expanded to include Code output with detailed Markdown mathematical formulations. Evaluation methodology was clarified with updated Agents and Tasks references. Benchmark results were updated from evaluation date 2026-06-22 to 2026-06-23 with revised scores for Correctness, Discoverability, Effectiveness, and Efficiency. Testing status was expanded with Agent Red-Teaming and marked security items as not completed. The Sigstore signature bundle was updated to reflect content changes. A blank line was added to SKILL.md for formatting.

Changes

Skill Evaluation Documentation Update

Layer / File(s)	Summary
Skill definition and evaluation setup skills/cuopt-numerical-optimization-formulation/skill-card.md	Use Case description rephrased to explicitly define parsing unstructured optimization problems into LP/MILP/QP formulations. Skill Output expanded to declare Code alongside Analysis output types with Markdown mathematical formulations. Evaluation Agents and Evaluation Tasks metadata revised to reference 3 internal NVSkills-Eval evaluation tasks focused on positive activation cases.
Evaluation results and testing status skills/cuopt-numerical-optimization-formulation/BENCHMARK.md, skills/cuopt-numerical-optimization-formulation/skill-card.md	Evaluation date updated to 2026-06-23. Benchmark results table updated with new percentages and uplift/loss deltas for Correctness, Discoverability, Effectiveness, and Efficiency dimensions across claude-code and codex. Testing Completed checklist expanded to include Agent Red-Teaming entry and mark Network Security and Product Security as not completed.
Signature verification update skills/cuopt-numerical-optimization-formulation/skill.oms.sig	Sigstore DSSE bundle JSON replaced with updated envelope payload reflecting new subject digests, predicate resource entries, and ignore-path configuration. Bundle structure and media types remain unchanged.
Documentation formatting skills/cuopt-numerical-optimization-formulation/SKILL.md	Blank line added before the first section heading for improved vertical spacing after front matter.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~18 minutes

Possibly related PRs

• NVIDIA/cuopt#1398: Both PRs update the cuopt-numerical-optimization-formulation skill documentation and evaluation results in SKILL.md, BENCHMARK.md, and skill-card.md in response to evaluation run changes.

Suggested reviewers

• tmckayus
• mlubin

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title accurately describes the main purpose of the PR: triggering CI evaluations for the numerical optimization skill via a whitespace change.
Description check	✅ Passed	The description directly explains the PR's intent—a trivial whitespace change to trigger CI evaluations—which aligns with the documented changes and objectives.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch ci/trigger-numerical-opt-evals

---

_{Comment @coderabbitai help to get the list of available commands.}

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

[coderabbitai]

Actionable comments posted: 2

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@skills/cuopt-numerical-optimization-formulation/skill-card.md`:
- Line 67: The Efficiency row contains a confusing delta value of (-0%) for the
third column which should be normalized to (0%) for consistency and clarity in
the benchmark table. Locate the Efficiency row that shows 72% (+45%) and update
the corresponding delta value from (-0%) to (0%) to maintain uniform formatting
across all delta values.
- Line 39: The discoverability claim on Line 45 states the skill should avoid
being used when irrelevant, but Line 39 only documents evaluation against
positive skill-activation cases. To fix this methodological gap, either add
negative test cases to the evaluation and update Line 39 to reflect the expanded
test set, or narrow the discoverability description on Line 45 to only claim
measurable behavior from positive-only testing (such as only claiming successful
activation in relevant contexts, without claiming avoidance of irrelevant use
cases).

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 0fb56caa-3319-4997-968c-aba5f1d07504

📥 Commits

Reviewing files that changed from the base of the PR and between e3cbefa and 8e700c8.

📒 Files selected for processing (3)

• skills/cuopt-numerical-optimization-formulation/BENCHMARK.md
• skills/cuopt-numerical-optimization-formulation/skill-card.md
• skills/cuopt-numerical-optimization-formulation/skill.oms.sig

✅ Files skipped from review due to trivial changes (1)

• skills/cuopt-numerical-optimization-formulation/BENCHMARK.md

[coderabbitai]

🎯 Functional Correctness | 🟠 Major | ⚡ Quick win

Discoverability is over-claimed with positive-only tasks.

Line 39 says all 3 evaluation tasks are positive activation cases, but discoverability (Line 45) also includes “avoid using it when irrelevant.” That irrelevant-case behavior is not measurable from positive-only tasks, so the discoverability result is methodologically incomplete.

Suggested doc fix

-Evaluated against 3 internal evaluation tasks (all positive skill-activation cases) via NVSkills-Eval external profile. <br>+Evaluated against 3 internal evaluation tasks (all positive skill-activation cases) via NVSkills-Eval external profile. In this run, Discoverability reflects relevant-case activation only (no negative-control tasks). <br>

📝 Committable suggestion

‼️IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.

Suggested change

	Evaluated against 3 internal evaluation tasks (all positive skill-activation cases) via NVSkills-Eval external profile. <br>
	Evaluated against 3 internal evaluation tasks (all positive skill-activation cases) via NVSkills-Eval external profile. In this run, Discoverability reflects relevant-case activation only (no negative-control tasks). <br>

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@skills/cuopt-numerical-optimization-formulation/skill-card.md` at line 39,
The discoverability claim on Line 45 states the skill should avoid being used
when irrelevant, but Line 39 only documents evaluation against positive
skill-activation cases. To fix this methodological gap, either add negative test
cases to the evaluation and update Line 39 to reflect the expanded test set, or
narrow the discoverability description on Line 45 to only claim measurable
behavior from positive-only testing (such as only claiming successful activation
in relevant contexts, without claiming avoidance of irrelevant use cases).

[coderabbitai]

📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick win

Normalize delta formatting to avoid -0%.

Line 67 shows 28% (-0%), which is confusing in a published benchmark table. Use 0% for zero deltas.

Suggested doc fix

-| Efficiency | 3 | 72% (+45%) | 28% (-0%) |+| Efficiency | 3 | 72% (+45%) | 28% (0%) |

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@skills/cuopt-numerical-optimization-formulation/skill-card.md` at line 67,
The Efficiency row contains a confusing delta value of (-0%) for the third
column which should be normalized to (0%) for consistency and clarity in the
benchmark table. Locate the Efficiency row that shows 72% (+45%) and update the
corresponding delta value from (-0%) to (0%) to maintain uniform formatting
across all delta values.