agent-neural-network

Agent skill for neural-network - invoke with $agent-neural-network

1.60x

Quality

30%

Does it follow best practices?

Impact

96%

1.60x

Average score across 3 eval scenarios

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/agent-neural-network/SKILL.md

Quality

Content

50%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.

The body is a well-structured agent-prompt with useful MCP call templates and a clear six-step workflow, but it reads as a persona/system prompt rather than a lean skill file: it over-explains known concepts, uses placeholder arguments instead of executable recipes, and lacks validation checkpoints and external reference structure.

Suggestions

Tighten the persona boilerplate ('You are a Flow Nexus Neural Network Agent, an expert in...') and the generic Quality standards / Advanced capabilities lists, which restate knowledge Claude already has.

Replace placeholder arguments (model_id: "model_id", user_id: "user_id", tier: "small") with concrete, executable examples and add the actual commands for checkpointing, validation, and deployment so the guidance is copy-paste ready.

Add explicit validation/feedback-loop steps to the workflow (e.g., validate training metrics before promoting a model, re-run on failure) and consider moving the API reference and architecture catalog into a referenced REFERENCE.md rather than inlining everything.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient with organized lists and code, but includes boilerplate 'You are a... Agent, an expert in...' framing and generic quality-standards/advanced-capabilities prose that restates concepts Claude already knows, matching the score-3 anchor 'Mostly efficient but includes some unnecessary explanation or could be tightened'.	3 / 5
Actionability	It provides MCP call examples with realistic config fields, but arguments like model_id: "model_id", user_id: "user_id", and tier: "small" are illustrative templates rather than complete executable recipes, and there are no concrete commands for the training/deployment lifecycle; fits score-3 'incomplete; missing key details'.	3 / 5
Workflow Clarity	The six-step ML workflow is sequenced (Problem Analysis through Deployment Management), but there are no validation checkpoints or feedback loops despite training/deployment being risky operations; per the rubric cap, a skill lacking validation on batch/destructive-style operations cannot score above 3, which fits the score-3 anchor.	3 / 5
Progressive Disclosure	The content is organized into labeled sections, but it is a monolithic single file with no references to separate files and no bundle present; a substantial API/toolkit reference that could live in separate files is inlined, fitting score-3 'content that should be separate is inline'.	3 / 5
	Total	12 / 20 Passed

Description

11%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.

The frontmatter description is a placeholder-grade label rather than a real capability statement: it names the domain but conveys no concrete actions, no natural trigger phrases, and no use-when guidance. It reads as auto-generated boilerplate rather than something a user would recognize as a trigger.

Suggestions

Replace the description with concrete actions, e.g. 'Train, deploy, and manage distributed neural networks across cloud sandboxes using Flow Nexus; handles architecture setup, distributed training, inference, and model lifecycle.'

Add an explicit 'Use when...' clause with natural user triggers such as 'Use when the user asks to train, fine-tune, or deploy a neural network, run distributed ML training, or manage model inference via Flow Nexus.'

Drop the 'Agent skill for ... - invoke with $...' boilerplate; it is invocation syntax, not a description of capability, and adds no trigger value.

Dimension	Reasoning	Score
Specificity	The description is 'Agent skill for neural-network - invoke with $agent-neural-network', which names only the domain via a generic 'Agent skill' label with zero concrete actions, matching the score-1 anchor 'no concrete actions; pure abstract language'.	1 / 5
Completeness	It states neither what the skill concretely does nor when to use it, matching the score-1 anchor 'Missing both what and when'; 'invoke with $agent-neural-network' is invocation syntax, not a use trigger.	1 / 5
Trigger Term Quality	The phrase contains 'neural-network' but no natural trigger phrases a user would say (e.g., 'train a model', 'deploy a neural network'); it sits between anchor 1 (no natural keywords) and anchor 3 (some relevant keywords), closer to 2 because the only keyword is the domain label.	2 / 5
Distinctiveness Conflict Risk	'Agent skill for neural-network' is very broad with high overlap risk against any ML-adjacent skill, fitting the score-2 anchor 'Very broad; high overlap risk with many similar skills'.	2 / 5
	Total	6 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: ruvnet/ruflo
Path: .agents/skills/agent-neural-network/SKILL.md
Commit: fa13ee4

Reviewed: about 4 hours 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.