Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add aws-neuron/neuron-agentic-development --skill neuron-framework-equivalencegit clone --depth 1 https://github.com/aws-neuron/neuron-agentic-developmentWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/aws-neuron/neuron-agentic-development/neuron-framework-equivalence)<a href="https://agentmods.dev/skills/aws-neuron/neuron-agentic-development/neuron-framework-equivalence"><img src="https://agentmods.dev/badge/skills/aws-neuron/neuron-agentic-development/neuron-framework-equivalence/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/aws-neuron/neuron-agentic-development/neuron-framework-equivalence"><img src="https://agentmods.dev/badge/skills/aws-neuron/neuron-agentic-development/neuron-framework-equivalence.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium analysis-evasion · line 1 Suspicious Unicode normalization or mixed-script contentFix: Review the flagged content for security risks. Ensure no credentials, secrets, or sensitive data are exposed.
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00100 | $0.03582 |
| Opus 5.5 | $0.00040 | $0.01433 |
| Sonnet 5.5 | $0.00020 | $0.00716 |
| Haiku 4.5 | $0.00010 | $0.00358 |
Grade A, and why
neuron-framework-equivalence scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 257 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Model Equivalence Framework
Verify and diagnose functional equivalence between a source (reference) and a target (ported) implementation of the same model.
Required Inputs
Before starting, collect these from the user. Ask for any missing ones.
| Input | Description | Example |
|---|---|---|
SOURCE_MODEL_PATH |
Path to source model weights (HF format) | /path/to/hf_models/Qwen3-0.6B |
COMPILED_MODEL_PATH |
Path to compiled target model | /path/to/neuron_models/Qwen3-0.6B |
TARGET_MODELING_FILE |
Path to target's modeling .py file | /path/to/modeling_qwen3.py |
TARGET_INNER_CLASS |
Inner model class (extends NeuronBaseModel) | NeuronQwen3Model |
TARGET_CAUSAL_CLASS |
ForCausalLM wrapper class | NeuronQwen3ForCausalLM |
TARGET_CONFIG_CLASS |
InferenceConfig class | Qwen3InferenceConfig |
VENV |
Path to Python venv with torch + neuronx | /opt/aws_neuronx_venv_pytorch_2_8_nxd_inference |
EXP_DIR |
Experiment output directory | agent_artifacts/equiv_qwen3 |
VLLM_NEURON_DIR |
vLLM-Neuron targets only. Project root of the vllm-neuron editable install |
/path/to/vllm-neuron |
Set SCRIPTS_DIR to the absolute path of this skill's scripts/ directory.
vLLM-Neuron Targets: Version Pin + PYTHONPATH
This skill's vllm_neuron adapter is pinned to vLLM 0.24.0 and the vLLM-Neuron plugin 0.24.0 line (two independently-versioned packages; the pins live in scripts/adapters/vllm_neuron.py as PINNED_VLLM_VERSION and PINNED_VLLM_NEURON_VERSION). When the target stack is vLLM-Neuron:
- Version check is automatic.
get_adapter()callsadapter.check_environment()immediately after construction. If the installedvllmorvllm-neuronplugin is not on the0.24line, orvllm_neuronis not importable, the run exits early with a clear message — no cryptic mid-stageAssertionError: Current vLLM config is not set. - Editable install must be on PYTHONPATH.
vllm_neuronis a local editable install, and vLLM's plugin entry point (vllm_neuron:register) imports it. PrependVLLM_NEURON_DIRtoPYTHONPATHon every stage command for vLLM-Neuron targets:
What ships with it
60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/adapter-contract.md 7.0 KB
- references/cpu-component-debugging.md 17 KB
- references/cpu-e2e-debugging.md 7.1 KB
- references/debug-orchestration.md 3.6 KB
- references/debugging-case-study-gptoss.md 7.8 KB
- references/device-component-debugging.md 8.1 KB
- references/device-e2e-debugging.md 7.0 KB
- references/dump-tensors.md 6.4 KB
- references/enable-model-run.md 4.2 KB
- references/equiv-concept.md 9.6 KB
- references/example_plots/negative_samples/bf16_3tensor_tp1_diff_qqplot_hist.png 99 KB
- references/example_plots/negative_samples/M2304_K256_N4096_seed42_d2a166b27b5e5073e755ff9ead80243e_debug_cpu32_gpu_trn_diff_qqplot_hist.png 101 KB
- references/example_plots/negative_samples/M256_K128_N4096_seed42_916d411da0ea543e2d2afe79671115bb_debug_cpu32_gpu_trn_diff_qqplot_hist.png 104 KB
- references/example_plots/negative_samples/Qwen3-8B-feedforward-module-mismatch-2.png 263 KB
- references/example_plots/negative_samples/Qwen3-8B-feedforward-module-mismatch.png 260 KB
- references/example_plots/negative_samples/rmsnorm.buggy_diff_qqplot_hist.png 71 KB
- references/example_plots/positive_samples/M2304_K1024_N4096_seed42_481f8ec28e6c2b4f0050e4dee2b5b550_debug_cpu32_cpu_trn_diff_qqplot_hist.png 82 KB
- references/example_plots/positive_samples/M2304_K1024_N4096_seed42_481f8ec28e6c2b4f0050e4dee2b5b550_debug_cpu32_gpu_trn_diff_qqplot_hist.png 84 KB
- references/example_plots/positive_samples/M2304_K256_N4096_seed42_d2a166b27b5e5073e755ff9ead80243e_debug_cpu32_cpu_trn_rel_diff_qqplot_hist.png 97 KB
- references/example_plots/positive_samples/M256_K128_N4096_seed42_916d411da0ea543e2d2afe79671115bb_debug_cpu32_cpu_trn_diff_qqplot_hist.png 100 KB
- references/example_plots/positive_samples/rmsnorm.fixed_diff_qqplot_hist.png 68 KB
- references/expected_structural_diffs.md 2.6 KB
- references/mapping_example.json 15 KB
- references/report-template.md 2.6 KB
- references/vllm-neuron-adaptation.md 7.7 KB
- scripts/adapters/__init__.py 3.0 KB runs code
- scripts/adapters/base.py 14 KB runs code
- scripts/adapters/nxdi.py 6.6 KB runs code
- scripts/adapters/vllm_neuron.py 42 KB runs code
- scripts/detect_class_divergence.py 9.5 KB runs code
- scripts/get_model_tree.py 22 KB runs code
- scripts/nxdi_compiled_loader.py 10 KB runs code
- scripts/run_calibration.py 3.7 KB runs code
- scripts/run_stage0.py 3.7 KB runs code
- scripts/run_stage1.py 5.7 KB runs code
- scripts/run_stage2.py 6.4 KB runs code
- scripts/run_stage3.py 4.9 KB runs code
- scripts/run_stage5.py 12 KB runs code
- scripts/run_stage6.py 6.0 KB runs code
- scripts/run_stage7.py 9.0 KB runs code
- scripts/run_teacher_forced_comparison.py 19 KB runs code
- scripts/stage0_scaffolding.py 16 KB runs code
- scripts/tensor_compare.py 15 KB runs code
- STAGE0.md 5.9 KB
- STAGE1.md 2.4 KB
- STAGE2.md 8.6 KB
- STAGE3.md 1.3 KB
- STAGE4.md 4.1 KB
- STAGE5.md 3.9 KB
- STAGE6.md 996 B
- STAGE7.md 2.3 KB
- templates/bias_restoration_template.py 4.3 KB runs code
- templates/conftest_template.py 2.4 KB runs code
- templates/conftest_vllm_template.py 1.7 KB runs code
- templates/diagnostic_forward_template.py 6.0 KB runs code
- templates/equiv_config_template.json 1.1 KB
- templates/fast_tp_equiv_test_template.py 9.9 KB runs code
- templates/neuron_device_validator_template.py 6.9 KB runs code
- templates/run_inference_cpu_template.py 9.9 KB runs code
- templates/test_template.py 2.6 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 257 lines · 100 tokens per session scan A 361dcc6e607a
neuron-framework-equivalence is a skill published in the GitHub repository aws-neuron/neuron-agentic-development (63 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 100 tokens to every session and 3,582 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
research-engineer
An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.
eval-dataset-design
A guide to creating reliable test sets for AI agents. Each evaluation task includes the request, the simulated user's behaviour, a reset starting state, and a success check that can be independently verified.
tika-eval-compare
Compare extracts from two Tika builds over a corpus to detect regressions in content, encoding, exceptions, and embedded-document handling. Use for "compare before/after extracts", "eval this change against the corpus".
cua-driver
Use Cua Driver for desktop or browser tasks that are awkward or unavailable through Bash/APIs, or when the user explicitly wants GUI interaction: app testing, visual bug reproduction, form filling, calendar entry, screenshots, and demo recording. Also covers Cua setup; not OpenAI Codex Computer Use or web research.
jetson-validate-image
Use after jetson-flash-image to run static BSP checks, on-target smoke/regression tests on a flashed DUT, or both. Not for build or flash steps. Triggers: validate bsp, on-target validation.
atmos-validation
Validate Atmos projects, components, arbitrary JSON Schema inputs, EditorConfig, and GitHub Actions; use affected-file selection and native CI annotations.