CtrlK
BlogDocsLog inGet started
Tessl Logo

cuopt-user-rules

Base rules for end users calling NVIDIA cuOpt (routing/LP/MILP/QP/install/server). Not for cuOpt internals — use cuopt-developer for those.

65

Quality

80%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./plugins/nvidia/skills/cuopt-user-rules/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

77%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The content is highly actionable with concrete commands, phrases, and a well-sequenced solve-and-verify workflow, but it is somewhat repetitive across its 'ask before acting' sections and keeps everything inline rather than splitting detail into reference files.

Suggestions

Consolidate the overlapping 'ask before assuming / running / installing' sections into a single decision table or one consolidated checkpoint to remove redundancy.

Move the package/installation detail (the interface→package table and verification commands) into a references file (e.g. INSTALL.md) and link to it from a leaner Check Environment section.

Trim the synthesized-data documentation block in 'Clarify Data Requirements' to a one-line directive plus an example, since the multi-step prose can be tightened.

DimensionReasoningScore

Conciseness

The body is mostly efficient behavioral rules but the 'ask before acting' theme recurs across several sections (Ask Before Assuming, Handle Incomplete Questions, Clarify Data Requirements, MUST Verify Understanding, Ask Before Running, Never Install Packages Automatically), creating repetition that could be tightened; it is not level 3 because not every token earns its place, and not level 1 because it does not pad with concepts Claude already knows.

2 / 3

Actionability

Concrete, copy-paste-ready guidance abounds — exact commands ("curl http://localhost:8000/cuopt/health", "pip install ..."), a package-by-interface table, and verbatim phrases to say ("Let me confirm I understand:..."); it is not level 2 because the instructions are executable and specific rather than pseudocode or abstract direction.

3 / 3

Workflow Clarity

The solve loop is explicitly sequenced with checkpoints — verify understanding first, formulate once, solve, sanity-check status/objective, and a mandatory post-correction check — matching the level-3 anchor of a clear sequence with validation and feedback loops; it is not level 2 because checkpoints are explicit, not implicit.

3 / 3

Progressive Disclosure

The single SKILL.md is well-organized into header-divided sections and a Resources section with one-level external links, but all ~215 lines are inline with no bundle files or split reference content; it is not level 3 because content is not appropriately split across files, and not level 1 because organization is clear and references are not deeply nested.

2 / 3

Total

10

/

12

Passed

Description

82%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific, well-triggered, and cleanly distinguished from the sibling developer skill, but it omits an explicit 'Use when...' trigger clause, which caps completeness at 2.

Suggestions

Add an explicit 'Use when...' clause, e.g. 'Use when helping end users call, install, or deploy NVIDIA cuOpt (routing/LP/MILP/QP) — for modifying cuOpt internals, use cuopt-developer.'

Consider pairing the technical abbreviations (LP/MILP/QP) with one natural-language phrasing such as 'linear and mixed-integer programming' so non-expert users still trigger the skill.

DimensionReasoningScore

Specificity

"calling NVIDIA cuOpt (routing/LP/MILP/QP/install/server)" lists multiple concrete capability areas rather than vague abstractions, matching the level-3 anchor of multiple specific concrete actions; it is not level 2 because the enumeration is broad and specific rather than naming only a domain and a few actions.

3 / 3

Completeness

"Base rules for end users calling NVIDIA cuOpt" answers the 'what' and implies the 'when', but there is no explicit "Use when..." clause; per the judging guidelines a missing explicit trigger clause caps completeness at 2, and it is not level 1 because the what is clearly stated and the when is implied.

2 / 3

Trigger Term Quality

Natural terms a cuOpt user would say appear directly — "NVIDIA cuOpt", "routing", "install", "server" — giving good coverage; it is not level 2 because the core product name and common task terms are present rather than only technical jargon or generic phrasing.

3 / 3

Distinctiveness Conflict Risk

"Not for cuOpt internals — use cuopt-developer for those" carves out a clear niche and actively redirects the overlapping case, making wrong-skill triggering unlikely; it is not level 2 because the disambiguation is explicit rather than merely somewhat specific.

3 / 3

Total

11

/

12

Passed

Validation

81%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation13 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

13

/

16

Passed

Repository
openai/plugins
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.