neural-training

Neural pattern training with SONA (Self-Optimizing Neural Architecture), MoE (Mixture of Experts), and EWC++ for knowledge consolidation. Use when: pattern learning, model optimization, knowledge transfer, adaptive routing. Skip when: simple tasks, no learning required, one-off operations.

Quality

80%

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 ./.agents/skills/neural-training/SKILL.md

Quality

Content

76%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 concise and highly actionable with ready-to-run commands and a clean component table. Its main weakness is workflow clarity: the training/consolidation pipeline lacks the validation checkpoints the rubric requires for batch operations that change model state.

Suggestions

Add a validation/verification checkpoint to the Intelligence Pipeline or Commands (e.g., run `npx claude-flow neural status` after `train`/`optimize` and only proceed when it reports success) to lift workflow_clarity above the batch-operation cap of 3.

Remove or relocate the marketing-style performance numbers in the Components table ("150x-12,500x faster", "2.49x-7.47x") to tighten conciseness, or move them to a references file.

Tie the abstract Pipeline steps (RETRIEVE/DISTILL/CONSOLIDATE) to the concrete commands below so each stage maps to an executable action.

Dimension	Reasoning	Score
Conciseness	The body is lean with tight tables and copy-paste command blocks, but marketing-style performance figures ("150x-12,500x faster", "2.49x-7.47x") are tokens Claude does not need and could be trimmed.	4 / 5
Actionability	Five concrete, copy-paste-ready `npx claude-flow neural ...` commands cover the common cases (train, status, patterns, predict, optimize), matching the fully-executable top anchor.	5 / 5
Workflow Clarity	The Intelligence Pipeline lists a RETRIEVE→JUDGE→DISTILL→CONSOLIDATE sequence, but training/consolidation are batch operations that mutate model state with no validation or verification checkpoints; per the rubric cap this cannot exceed 3.	3 / 5
Progressive Disclosure	A single well-organized file with clear sections (Purpose, When to Trigger, Pipeline, Components, Commands, Best Practices) and no bundle files present; good structure, though slightly above the sub-50-line simple-skill threshold.	4 / 5
	Total	16 / 20 Passed

Description

83%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 well-formed description that clearly states what the skill does and when to use it, with concrete trigger and skip clauses. Voice is appropriately third-person and buzzwords are glossed with expansions, though a few trigger terms remain somewhat technical.

Dimension	Reasoning	Score
Specificity	Names the domain ("Neural pattern training") and several concrete capabilities via named systems ("SONA ... MoE ... EWC++ for knowledge consolidation", "knowledge transfer", "adaptive routing"); minor gaps in coverage keep it below a 5.	4 / 5
Completeness	It explicitly answers both "what" (neural pattern training with SONA/MoE/EWC++ for consolidation) and "when" ("Use when: ..." plus "Skip when: ...") with concrete trigger phrases, matching the top anchor.	5 / 5
Trigger Term Quality	The "Use when" clause supplies natural phrases ("pattern learning, model optimization, knowledge transfer, adaptive routing") that users would plausibly say, though terms like "adaptive routing" lean technical and a few common synonyms are missing.	4 / 5
Distinctiveness Conflict Risk	The named-system niche (SONA/MoE/EWC++) and explicit triggers make it mostly distinct, though broad terms like "model optimization" and "knowledge transfer" carry minor overlap risk with other ML skills.	4 / 5
	Total	17 / 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/neural-training/SKILL.md
Commit: b14c79e

Reviewed: 3 days 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.