SKILLEMALL.ai
GitHub ↗ RU Sign in

CD nemo-automodel-model-onboarding

Guide for onboarding new model architectures into NeMo AutoModel, including architecture discovery, implementation patterns, registration, and validation. Use when adding or modifying model architecture support in NeMo AutoModel, such as LLM/VLM/MoE model files, custom layers, state-dict adapters, registry entries, Hugging Face config mapping, or capability flags.

NVIDIA-NeMo/Automodel Agent Skills author: NVIDIA-NeMo 8 files body ≈ 5 572 tokens Open the sourcegithub.com↗ analyzed 8 h ago

Guide for onboarding new model architectures into NeMo AutoModel, including architecture discovery, implementation patterns, registration, and validation.

As a process D 36/100 · Unfinished process — References files that are not bundled: ../nemo-automodel-recipe-development/SKILL.md, examples/llm_finetune/<name>/, examples/vlm_finetune/<name>/

AnalyzerAI and agentsWriting and documentstype and topics are labelled automatically from the skill text
Runs in: Claude / Claude Code ChatGPT / Codex Gemini CLI Cursor GitHub Copilot / VS Code Qwen Code Kimi Code CLI Z.ai (GLM) DeepSeek (Deep Code) Mistral Vibe LangChain (deepagents) OpenClaw Goose OpenCode
JSON
Technical rating
C
83/100
safety, quality, tests
Safety 60%
97
Quality 40%
63
Run on models
none yet
Process rating
D
36/100
Unfinished process
References files that are not bundled: ../nemo-automodel-recipe-development/SKILL.md, examples/llm_finetune/<name>/, examples/vlm_finetune/<name>/
Tools and files w 18
0
Result and completion w 14
0
Inputs and preconditions w 11
0
the three weakest of ten parameters · all ten

How to improve

  1. The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
  2. The text references files that are not there: add them or drop the references.
For the model run — optional
  • A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.

Guard findings · 3

✓ No critical or high findings

Medium and low: 3
  • low Secrets in code secret-high-entropy-token SKILL.md:90
    High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)
    - `architectures` -- determines the class name and registration key (e.g., `"LlamaForCausalLM"`, `"Qwen…lLM"`, `"Mist…ion"`)
    quoted
  • low Secrets in code secret-high-entropy-token vlm-patterns.md:6
    High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)
    - `components/models/mistral4/model.py` -- `Mist…ion` (Pixtral vision + MoE text)
    quoted
  • low Secrets in code secret-high-entropy-token vlm-patterns.md:8
    High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)
    - `components/models/kimi_k25_vl/model.py` -- `Kimi…ion`
    quoted

Files scanned: 7. Evidence is masked. Grey chips explain why severity was lowered.

Against the Agent Skills spec

  • warning body-long SKILL.md body ≈ 5572 tokens (recommended < 5000); move details to references/
  • warning missing-ref reference to a missing file: ../nemo-automodel-recipe-development/SKILL.md
  • warning missing-ref reference to a missing file: examples/llm_finetune/<name>/
  • warning missing-ref reference to a missing file: examples/vlm_finetune/<name>/

Process rating: all ten parameters 36/100

Will not run. References files that are not bundled: ../nemo-automodel-recipe-development/SKILL.md, examples/llm_finetune/<name>/, examples/vlm_finetune/<name>/
  • 0Tools and files. 3 referenced file(s) missing: ../nemo-automodel-recipe-development/SKILL.md, examples/llm_finetune/<name>/, examples/vlm_finetune/<name>/
  • 0Result and completion. Does not say what the result is
  • 0Inputs and preconditions. Does not say what the process needs to start
  • 0Progress reporting. Says nothing while it works
  • 50When it triggers. No condition that starts the skill
  • 50Failures and branches. 0 branches, has a failure section
  • 60Steps. 95 steps, 6 vague phrases
  • 70Execution cost. Instruction body is 5572 tokens
  • 100Consistency. Name and required fields are in place
  • 100Running it twice. Mutating operations check current state
  • low 12 top-level sections: this looks like several domains in one skill
  • low The response is described with custom markup (7 tags): a typed call is more reliable
  • medium 3 test cases, all positive: not one "should refuse" or "should ask first"

Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.

Quality signals

  • +5Description has no quoted example phrases that should trigger the skill
  • +4Description does not say when NOT to use the skill (false activations)
  • +3Output format is not stated: the model decides each time
  • +2Single-language instructions
  • +3Description length 366: enough signal without eating the budget
  • +4Structure: 30 headings
  • +3Step-by-step instructions: 95 items
  • +4Has examples (7 code blocks)
  • +1License stated

Quality base 70; lint remarks subtract, signals add up to 100. Result: 63.