chore: trigger NVSkills CI for cuopt-user-rules

[ramakrishnap-nv]

Trivial blank-line addition after SKILL.md frontmatter to trigger NVSkills CI — fetch updated skill card and benchmark for cuopt-user-rules.

[ramakrishnap-nv]

/nvskills-ci

[coderabbitai]

📝 Walkthrough

Walkthrough

Updated the cuOpt user rules skill metadata, evaluation cases, publication benchmark report, document spacing, and Sigstore/in-toto attestation.

Changes

cuOpt user rules publication refresh

Layer / File(s)	Summary
Skill metadata and evaluation skills/cuopt-user-rules/skill-card.md, skills/cuopt-user-rules/evals/evals.json	Expanded supported interfaces and evaluation metadata. Updated metrics, signals, agent-specific results, and added six evaluation cases.
Publication benchmark skills/cuopt-user-rules/BENCHMARK.md	Replaced the previous evaluation report with August publication metadata, score comparisons, tier statuses, findings, scoring rules, and freshness criteria.
Documentation and attestation skills/cuopt-user-rules/SKILL.md, skills/cuopt-user-rules/skill.oms.sig	Added heading spacing and updated attestation digests, resource ordering, and DSSE signature data.

Estimated code review effort: 3 (Moderate) | ~20 minutes

Possibly related PRs

• NVIDIA/cuopt#1676: Applies the same NVSkills evaluation, reporting, skill-card, attestation, and evaluation-case updates to another cuOpt skill.

Suggested reviewers: tmckayus, iroy30

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly identifies the CI-triggering change for the cuopt-user-rules skill.
Description check	✅ Passed	The description accurately explains the blank-line change and the resulting NVSkills CI updates.
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.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch chore/nvskills-ci-cuopt-user-rules

---

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

[github-actions]

CI Test Summary

⏭️ All 5 test job(s) skipped.

[ramakrishnap-nv]

/nvskills-ci

[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-user-rules/skill-card.md`:
- Line 74: Correct the Claude Code effectiveness uplift from “+26 points” to
“+27 points” in skills/cuopt-user-rules/skill-card.md lines 74-74 and
skills/cuopt-user-rules/BENCHMARK.md lines 41-41, keeping the surrounding
benchmark values unchanged.
- Around line 18-19: Update the credential metadata in the cuOpt skill card so
it explicitly declares that an NGC API key is required for registry
authentication. Replace the unspecified API-key requirement and “None
identified” credential type with the appropriate NGC API key information,
preserving the existing metadata format.

🪄 Autofix

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: d44c87df-f0c1-4ba5-95e8-86b1f3385dc5

📥 Commits

Reviewing files that changed from the base of the PR and between 24e086e and a8d1cd0.

📒 Files selected for processing (3)

• skills/cuopt-user-rules/BENCHMARK.md
• skills/cuopt-user-rules/skill-card.md
• skills/cuopt-user-rules/skill.oms.sig

[coderabbitai]

🗄️ Data Integrity & Integration | 🟠 Major | ⚡ Quick win

Declare the NGC API key requirement.

Lines 18-19 state that no credential is identified. The cuOpt quick-start requires an NGC API key for registry authentication. Update the credential metadata so consumers do not omit a required credential.

Proposed correction

-**Requires API Key or External Credential:** [Not Specified] <br>
-**Credential Type(s):** [None identified] <br>
+**Requires API Key or External Credential:** [Yes] <br>
+**Credential Type(s):** [NGC API key] <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

	**Requires API Key or External Credential:** [Not Specified] <br>
	**Credential Type(s):** [None identified] <br>
	**Requires API Key or External Credential:** [Yes] <br>
	**Credential Type(s):** [NGC API key] <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-user-rules/skill-card.md` around lines 18 - 19, Update the
credential metadata in the cuOpt skill card so it explicitly declares that an
NGC API key is required for registry authentication. Replace the unspecified
API-key requirement and “None identified” credential type with the appropriate
NGC API key information, preserving the existing metadata format.

[ramakrishnap-nv]

/ok to test a8d1cd0

[ramakrishnap-nv]

/ok to test ca45cea

[ramakrishnap-nv]

/ok to test cc38f1b

[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-user-rules/evals/evals.json`:
- Around line 74-83: The evaluation input is missing the partial cuOpt script,
user-defined names and structures, and the incorrect constraint needed for the
requested behavior. Update the `question` fixture to include a representative
incomplete script with a concrete incorrect constraint, or revise
`expected_behavior` and `ground_truth` to require requesting the missing code
before making changes.
- Around line 18-97: Add a partial cuOpt script and constraint section to the
fixture for user-rules-eval-006, or update its expected behavior to require
requesting the missing code before editing. Refresh the evals.json digest in
skill.oms.sig after the final skill changes and rerun NVSkills CI. If the
contributor-authored merge commits remain in the PR, add the required
Signed-off-by trailers.

🪄 Autofix

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: daad2d48-ad36-40ec-9e37-79b70ba130f0

📥 Commits

Reviewing files that changed from the base of the PR and between a8d1cd0 and 057e8e7.

📒 Files selected for processing (1)

• skills/cuopt-user-rules/evals/evals.json

[coderabbitai]

📐 Maintainability & Code Quality | 🟠 Major | ⚡ Quick win

🧩 Analysis chain

🏁 Script executed:

#!/bin/bash
set -euo pipefail

# Verify target branch, CI state, and commit sign-offs for the current PR.
gh pr view --json baseRefName,statusCheckRollup,commits \
  --jq '{baseRefName, statusCheckRollup, commits: [.commits[] | {oid, messageHeadline, messageBody}]}'

# Inspect commits that changed this skill and its attestation history.
git log --format='%H%n%B%n---' origin/main..HEAD
git log --follow --format='%H %s' -- skills/cuopt-user-rules/skill.oms.sig

Repository: NVIDIA/cuopt

Length of output: 223

---

🏁 Script executed:

#!/bin/bash
set -u

printf '%s\n' '--- repository state ---'
git status --short --branch
git branch --all --no-color
git remote -v

printf '%s\n' '--- target files ---'
git ls-files 'skills/cuopt-user-rules/*' | sed -n '1,120p'
printf '%s\n' '--- evaluation excerpt ---'
nl -ba skills/cuopt-user-rules/evals/evals.json | sed -n '1,115p'

printf '%s\n' '--- skill validation artifacts ---'
find skills/cuopt-user-rules -maxdepth 2 -type f -printf '%p\n' | sort | sed -n '1,120p'

printf '%s\n' '--- recent commits affecting the evaluation and signature ---'
git log --all --oneline --decorate -20 -- skills/cuopt-user-rules/evals/evals.json skills/cuopt-user-rules/skill.oms.sig
printf '%s\n' '--- signature history ---'
git log --all --follow --format='%H %s' -- skills/cuopt-user-rules/skill.oms.sig | sed -n '1,80p'

Repository: NVIDIA/cuopt

Length of output: 2059

---

🏁 Script executed:

#!/bin/bash
set -u

git status --short --branch
git branch --all --no-color
git remote -v

nl -ba skills/cuopt-user-rules/evals/evals.json | sed -n '1,115p'
find skills/cuopt-user-rules -maxdepth 2 -type f -printf '%p\n' | sort

git log --all --oneline --decorate -20 -- skills/cuopt-user-rules/evals/evals.json skills/cuopt-user-rules/skill.oms.sig
git log --all --follow --format='%H %s' -- skills/cuopt-user-rules/skill.oms.sig | sed -n '1,80p'

Repository: NVIDIA/cuopt

Length of output: 1675

---

🏁 Script executed:

#!/bin/bash
set -u

printf '%s\n' '--- evaluation records ---'
sed -n '1,115p' skills/cuopt-user-rules/evals/evals.json
python3 - <<'PY'
import json
from pathlib import Path
data = json.loads(Path("skills/cuopt-user-rules/evals/evals.json").read_text())
for item in data:
    print(json.dumps({
        "id": item.get("id"),
        "question": item.get("question"),
        "expected_script": item.get("expected_script"),
    }, ensure_ascii=False))
PY

printf '%s\n' '--- current revision and ancestry ---'
git show -s --format='HEAD=%H%n%B' HEAD
git rev-parse origin/main
git merge-base --is-ancestor origin/main HEAD
printf 'origin/main ancestor of HEAD: %s\n' "$?"
git rev-list --count origin/main..HEAD
git log --format='%H%x09%s%x09%b' origin/main..HEAD

printf '%s\n' '--- signature file and introducing commit ---'
git show HEAD:skills/cuopt-user-rules/skill.oms.sig
git show --stat --oneline a8d1cd0a
git show --format=fuller --no-ext-diff --unified=0 a8d1cd0a -- skills/cuopt-user-rules/skill.oms.sig | sed -n '1,100p'

printf '%s\n' '--- commit checks for current revision ---'
gh api repos/NVIDIA/cuopt/commits/057e8e73/check-runs \
  --jq '.check_runs[] | {name, status, conclusion, completed_at, html_url}'
gh api repos/NVIDIA/cuopt/commits/057e8e73/status \
  --jq '.statuses[] | {context, state, description, target_url}'

Repository: NVIDIA/cuopt

Length of output: 29344

---

🏁 Script executed:

#!/bin/bash
set -u

sed -n '1,115p' skills/cuopt-user-rules/evals/evals.json

python3 - <<'PY'
import json
from pathlib import Path
data = json.loads(Path("skills/cuopt-user-rules/evals/evals.json").read_text())
for item in data:
    print(json.dumps({
        "id": item.get("id"),
        "question": item.get("question"),
        "expected_script": item.get("expected_script"),
    }, ensure_ascii=False))
PY

git show -s --format='HEAD=%H%n%B' HEAD
git rev-parse origin/main
git merge-base --is-ancestor origin/main HEAD
printf 'origin/main ancestor of HEAD: %s\n' "$?"
git rev-list --count origin/main..HEAD
git log --format='%H%x09%s%x09%b' origin/main..HEAD

git show HEAD:skills/cuopt-user-rules/skill.oms.sig
git show --stat --oneline a8d1cd0a
git show --format=fuller --no-ext-diff --unified=0 a8d1cd0a -- skills/cuopt-user-rules/skill.oms.sig | sed -n '1,100p'

gh api repos/NVIDIA/cuopt/commits/057e8e73/check-runs \
  --jq '.check_runs[] | {name, status, conclusion, completed_at, html_url}'
gh api repos/NVIDIA/cuopt/commits/057e8e73/status \
  --jq '.statuses[] | {context, state, description, target_url}'

Repository: NVIDIA/cuopt

Length of output: 29190

---

🏁 Script executed:

#!/bin/bash
set -u

printf '%s\n' '--- signature resource digests versus current files ---'
python3 - <<'PY'
import base64, hashlib, json
from pathlib import Path

sig = json.loads(Path("skills/cuopt-user-rules/skill.oms.sig").read_text())
payload = json.loads(base64.b64decode(sig["dsseEnvelope"]["payload"]))
for resource in payload["predicate"]["resources"]:
    name = resource["name"]
    path = Path("skills/cuopt-user-rules") / name
    digest = hashlib.sha256(path.read_bytes()).hexdigest() if path.is_file() else "MISSING"
    print(f"{name}: signed={resource['digest']} current={digest} match={digest == resource['digest']}")
print("subject:", payload["subject"])
PY

printf '%s\n' '--- commits in the current change range and DCO trailers ---'
for commit in $(git rev-list origin/main..HEAD); do
  git show -s --format='%H%n%an <%ae>%n%s%n%(trailers:key=Signed-off-by,separator=|)%n---' "$commit"
done

printf '%s\n' '--- pull requests associated with the current commit ---'
gh api repos/NVIDIA/cuopt/commits/057e8e73ef7db03784a62e0719cf7239930646dc/pulls \
  --jq '.[] | {number, title, state, base: .base.ref, head: .head.ref, head_repo: .head.repo.full_name}'

Repository: NVIDIA/cuopt

Length of output: 2401

---

🏁 Script executed:

#!/bin/bash
set -u

python3 - <<'PY'
import base64, hashlib, json
from pathlib import Path

sig = json.loads(Path("skills/cuopt-user-rules/skill.oms.sig").read_text())
payload = json.loads(base64.b64decode(sig["dsseEnvelope"]["payload"]))
for resource in payload["predicate"]["resources"]:
    name = resource["name"]
    path = Path("skills/cuopt-user-rules") / name
    digest = hashlib.sha256(path.read_bytes()).hexdigest() if path.is_file() else "MISSING"
    print(f"{name}: signed={resource['digest']} current={digest} match={digest == resource['digest']}")
print("subject:", payload["subject"])
PY

for commit in $(git rev-list origin/main..HEAD); do
  git show -s --format='%H%n%an <%ae>%n%s%n%(trailers:key=Signed-off-by,separator=|)%n---' "$commit"
done

gh api repos/NVIDIA/cuopt/commits/057e8e73ef7db03784a62e0719cf7239930646dc/pulls \
  --jq '.[] | {number, title, state, base: .base.ref, head: .head.ref, head_repo: .head.repo.full_name}'

Repository: NVIDIA/cuopt

Length of output: 2227

---

Add the missing fixture input and refresh NVSkills validation.

• user-rules-eval-006-extend-not-rewrite has no script or constraint section, so its expected behavior cannot be evaluated. Include the partial script, or require the agent to request the missing code.
• skill.oms.sig contains a stale evals/evals.json digest, and require-nvskills-ci is failing. Re-sign the final skill contents and rerun NVSkills CI.
• Three contributor-authored merge commits in the change range lack Signed-off-by trailers. Add the required DCO sign-offs if these commits remain in the PR.

🤖 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-user-rules/evals/evals.json` around lines 18 - 97, Add a partial
cuOpt script and constraint section to the fixture for user-rules-eval-006, or
update its expected behavior to require requesting the missing code before
editing. Refresh the evals.json digest in skill.oms.sig after the final skill
changes and rerun NVSkills CI. If the contributor-authored merge commits remain
in the PR, add the required Signed-off-by trailers.

Source: Coding guidelines

[coderabbitai]

🗄️ Data Integrity & Integration | 🟠 Major | ⚡ Quick win

Add the partial script and the incorrect constraint to the test input.

The question provides no script, variable names, structures, or constraint section. expected_script is also null. A compliant agent must request the missing code. It cannot extend the script or make the required targeted constraint fix.

This evaluation can mark compliant behavior as a failure. Include a representative partial script with a concrete incorrect constraint in question, or change the expected behavior to request the missing input.

🤖 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-user-rules/evals/evals.json` around lines 74 - 83, The
evaluation input is missing the partial cuOpt script, user-defined names and
structures, and the incorrect constraint needed for the requested behavior.
Update the `question` fixture to include a representative incomplete script with
a concrete incorrect constraint, or revise `expected_behavior` and
`ground_truth` to require requesting the missing code before making changes.

[ramakrishnap-nv]

/nvskills-ci

[coderabbitai]

Actionable comments posted: 3

🤖 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-user-rules/BENCHMARK.md`:
- Line 37: Correct the uplift values in the benchmark table and the
corresponding entries in skill-card.md: update Codex Correctness to +12 and
Claude Code Efficiency to +47, preserving the existing table formatting and all
other scores.

In `@skills/cuopt-user-rules/skill-card.md`:
- Line 70: Correct the uplift values in the comparison table: set Codex
Correctness to +12 (83−71) and Claude Code Efficiency to +47 (72−25). Apply the
same corrections to the corresponding entries in BENCHMARK.md, preserving all
other values and formatting.
- Around line 70-75: Correct the repeated uplift values in both publication
artifacts: update Codex Correctness from +11 to +12 and Claude Code Efficiency
from +46 to +47 in skills/cuopt-user-rules/skill-card.md lines 70-75 and
skills/cuopt-user-rules/BENCHMARK.md lines 37-42.

🪄 Autofix

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: e1f4be9f-68de-4257-9284-0b073e3b8cf7

📥 Commits

Reviewing files that changed from the base of the PR and between 057e8e7 and ec57649.

📒 Files selected for processing (3)

• skills/cuopt-user-rules/BENCHMARK.md
• skills/cuopt-user-rules/skill-card.md
• skills/cuopt-user-rules/skill.oms.sig

🚧 Files skipped from review as they are similar to previous changes (1)

• skills/cuopt-user-rules/skill.oms.sig

[coderabbitai]

🎯 Functional Correctness | 🟡 Minor | ⚡ Quick win

Correct the two uplift calculations.

The table defines uplift as skill score - baseline score.

• Codex Correctness is 83 - 71 = 12, not +11.
• Claude Code Efficiency is 72 - 25 = 47, not +46.

Apply the same corrections in skills/cuopt-user-rules/skill-card.md.

Also applies to: 39-39, 42-42

🤖 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-user-rules/BENCHMARK.md` at line 37, Correct the uplift values
in the benchmark table and the corresponding entries in skill-card.md: update
Codex Correctness to +12 and Claude Code Efficiency to +47, preserving the
existing table formatting and all other scores.

[coderabbitai]

🎯 Functional Correctness | 🟡 Minor | ⚡ Quick win

Correct the two uplift calculations.

The table defines uplift as skill score - baseline score.

• Codex Correctness is 83 - 71 = 12, not +11.
• Claude Code Efficiency is 72 - 25 = 47, not +46.

Apply the same corrections in skills/cuopt-user-rules/BENCHMARK.md.

Also applies to: 72-72, 75-75

🤖 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-user-rules/skill-card.md` at line 70, Correct the uplift values
in the comparison table: set Codex Correctness to +12 (83−71) and Claude Code
Efficiency to +47 (72−25). Apply the same corrections to the corresponding
entries in BENCHMARK.md, preserving all other values and formatting.

[coderabbitai]

🎯 Functional Correctness | 🟡 Minor | ⚡ Quick win

Correct the repeated uplift arithmetic in both publication artifacts.

Both tables define uplift as skill score minus baseline, but repeat two incorrect differences.

• skills/cuopt-user-rules/skill-card.md#L70-L75: change Codex Correctness from +11 to +12, and Claude Code Efficiency from +46 to +47.
• skills/cuopt-user-rules/BENCHMARK.md#L37-L42: make the same two corrections.

📍 Affects 2 files

• skills/cuopt-user-rules/skill-card.md#L70-L75 (this comment)
• skills/cuopt-user-rules/BENCHMARK.md#L37-L42

🤖 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-user-rules/skill-card.md` around lines 70 - 75, Correct the
repeated uplift values in both publication artifacts: update Codex Correctness
from +11 to +12 and Claude Code Efficiency from +46 to +47 in
skills/cuopt-user-rules/skill-card.md lines 70-75 and
skills/cuopt-user-rules/BENCHMARK.md lines 37-42.

[ramakrishnap-nv]

/ok to test ec57649

[ramakrishnap-nv]

/merge