CtrlK
BlogDocsLog inGet started
Tessl Logo

monitor

Monitor submitted jobs (PTQ, evaluation, deployment) on SLURM clusters. Use when the user asks "check job status", "is my job done", "monitor my evaluation", "what's the status of the PTQ", "check on job <slurm_job_id>", or after any skill submits a long-running job. Also triggers on "nel status", "squeue", or any request to check progress of a previously submitted job.

77

Quality

96%

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

SKILL.md
Quality
Evals
Security

Quality

Content

92%Weight 40%Scale 1-3

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

A dense, actionable skill body with executable code, clear sequenced workflows, and explicit validation checkpoints. The main gap is progressive disclosure: all logic is inline with no reference or script bundle to offload detail.

Suggestions

Move the bash helper functions (extract_nel_state, extract_slurm_state, is_*_terminal) into a scripts/ file referenced from the body, so the SKILL.md overview stays lean and the helpers are reusable.

Consider splitting the per-job-type check sections (NEL, launcher, raw SLURM) into a references/ doc signaled from a short overview, enabling one-level-deep progressive disclosure.

Move the terminal-state vocabulary lists and discovery fallback steps into a reference file to further reduce inline bulk while keeping them discoverable.

DimensionReasoningScore

Conciseness

The body assumes Claude knows SLURM and avoids explaining basic concepts, leaning on dense code blocks and purposeful directives; the vocabulary-mixing warning and registry rationale are non-obvious domain knowledge that earns their tokens. It stays above level 2 because the prose is tight rather than padded with concepts Claude already knows.

3 / 3

Actionability

It provides executable bash helpers (extract_nel_state, is_nel_terminal, extract_slurm_state, is_slurm_terminal) and concrete commands (nel status <id>, sacct -j $jid -X --format=State, ssh "$host" ...) that are copy-paste ready; above level 2 because the code is complete and runnable, not pseudocode.

3 / 3

Workflow Clarity

"On Job Submission" and "On Monitor Event / Status Check" are explicitly numbered with validation checkpoints (compare against last_status, detect terminal states, remove completed jobs) and failure feedback loops ("On failure, diagnose immediately"), matching the level-3 anchor; above level 2 because validation and recovery steps are explicit, not implicit.

3 / 3

Progressive Disclosure

Sections are well-organized with clear headers, but everything lives inline in SKILL.md with no bundle files or one-level-deep references, so content that could be split (e.g. the per-job-type check helpers into a script) is inline. It does not reach level 3 because there is no appropriately split, well-signaled reference structure.

2 / 3

Total

11

/

12

Passed

Description

100%Weight 40%Scale 1-3

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

A strong, well-targeted description that clearly states the capability and gives rich, natural trigger phrases for when to use it. It is specific to the SLURM job-monitoring niche with low conflict risk.

DimensionReasoningScore

Specificity

"Monitor submitted jobs (PTQ, evaluation, deployment) on SLURM clusters" names the domain plus multiple concrete job categories (PTQ, evaluation, deployment, raw SLURM), matching the anchor for listing multiple specific concrete targets; it is above level 2 because the enumerated job types are concrete rather than a single vague action.

3 / 3

Completeness

It answers both what ("Monitor submitted jobs ... on SLURM clusters") and when ("Use when the user asks ...", "Also triggers on ...") with explicit trigger guidance, satisfying the level-3 anchor; not capped at 2 because the 'Use when' clause is present and explicit.

3 / 3

Trigger Term Quality

It includes natural phrases users would say ("check job status", "is my job done", "monitor my evaluation", "what's the status of the PTQ") plus command-level triggers ("nel status", "squeue"), giving good coverage of how users actually request this; not level 2 because common natural variations are present rather than only jargon.

3 / 3

Distinctiveness Conflict Risk

The SLURM/nel job-monitoring niche with tool-specific triggers (nel status, squeue, slurm_job_id) is clearly distinct and unlikely to fire for unrelated skills; above level 2 because the triggers are specific to this domain rather than overlapping generic terms.

3 / 3

Total

12

/

12

Passed

Validation

100%

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

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
NVIDIA/Model-Optimizer
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.