tensorpool-gpu-cloud

Safely inspect and operate TensorPool GPU clusters and jobs using the current tp CLI, with explicit approval before any billable or destructive action.

Quality

75%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./backend/cli/skills/cloud-compute/tensorpool/SKILL.md

Quality

Content

85%Weight 40%Scale 1-5

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

A tight, provider-specific policy document: concrete read commands, a well-sequenced approval-gated workflow with explicit error reconciliation, and correct deferral of volatile data (versions, pricing, inventory) to live inspection. The only real slack is slight redundancy in the credential-boundary section and the absence of the (deliberately gated) mutation command forms.

Dimension	Reasoning	Score
Conciseness	The body is lean and provider-specific throughout — every line carries TensorPool/OpenScience policy Claude would not otherwise know, and it deliberately avoids stale version data ('install and inspect the CLI rather than guessing its version'). Not 5 because the credential-boundary section repeats its core restriction across three bullets ('The saved key is admitted only when...', 'cannot be used by provider_compute', 'do not ask the user to weaken this boundary'), which could be tightened.	4 / 5
Actionability	Gives concrete, executable commands (`python -m pip install tensorpool`, `tp --no-input me`, `tp cluster list`, `tp job list`, `tp storage list`) plus specific UI paths (Customize > Compute > TensorPool / SSH) and the TENSORPOOL_KEY contract. Not 5 because the mutation commands (cluster create, job submit) are policy-gated without showing their current form or a `--help` lookup example for them, leaving a minor gap.	4 / 5
Workflow Clarity	The 7-step Operating policy is a clear sequence with explicit validation checkpoints (approval before every mutation, monitor to a terminal state, verify outputs before cleanup, confirm before deleting storage) and an explicit error-recovery feedback loop ('Never retry a billable mutation blindly after a timeout—first list resources and reconcile whether the first request succeeded'). This matches the top anchor: sequenced steps, explicit validation, and recovery guidance for destructive/billable operations.	5 / 5
Progressive Disclosure	The skill is self-contained at ~55 lines with no bundle files (references/, scripts/, assets/ are absent) and no content that warrants splitting; sections are well-organized and the 'Sources of truth' section points one level deep to live external docs rather than duplicating them. Per the simple-skill guideline this earns the top score on well-organized sections alone.	5 / 5
	Total	18 / 20 Passed

Description

65%Weight 40%Scale 1-5

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 distinct, moderately specific description whose main weakness is the missing 'Use when' trigger clause and a generic 'operate' verb that under-sells the concrete workflows covered. Adding an explicit trigger sentence and naming the key operations (create clusters, submit/monitor jobs, manage storage) would lift both completeness and specificity.

Suggestions

Add an explicit trigger clause, e.g. 'Use when the user mentions TensorPool, GPU clusters, GPU cloud instances, or running/monitoring GPU jobs.'

Replace the generic 'operate' with the concrete actions covered, such as creating clusters, submitting and monitoring jobs, and managing billable storage.

Include one or two natural synonyms users might say (e.g. 'GPU cloud', 'instances') to broaden trigger coverage.

Dimension	Reasoning	Score
Specificity	Names the domain ("TensorPool GPU clusters and jobs") and two actions ("inspect and operate") plus a policy qualifier ("explicit approval before any billable or destructive action"), but does not enumerate the concrete operations (cluster creation, job submission, storage) — closest to 'names domain and 1-2 concrete actions'. Not 4 because 'operate' is generic rather than a list of several specific actions; not 2 because the tp CLI and the approval boundary make the actions more concrete than 'minimal or generic'.	3 / 5
Completeness	The 'what' is clear (safely inspect and operate TensorPool GPU clusters/jobs via the current tp CLI with approval gating), but there is no 'Use when...' clause or equivalent explicit trigger guidance, which caps completeness at 3 per the judging guidelines. Not 4 because 'when' is entirely absent rather than merely imprecise.	3 / 5
Trigger Term Quality	Natural terms a user would say are present: 'TensorPool', 'GPU clusters', 'jobs', 'tp CLI'. Missing a few common variations such as 'GPU cloud', 'instances', or 'SSH', so it is 'good keyword coverage; a few natural terms missing' rather than comprehensive coverage with synonyms.	4 / 5
Distinctiveness Conflict Risk	The description is anchored to a single provider ('TensorPool ... using the current tp CLI'), giving it a clear niche with distinct triggers and minimal conflict risk with any other skill.	5 / 5
	Total	15 / 20 Passed

Validation

93%

Warnings & errors only

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

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: synthetic-sciences/openscience
Path: backend/cli/skills/cloud-compute/tensorpool/SKILL.md
Commit: 4082a2e

Reviewed: 1 day ago

Table of Contents

Discovery Implementation Validation

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.