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 agentmods add agents/aws-neuron/neuron-agentic-development/neuron-nki-profile-analysis-agentgit 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/agents/aws-neuron/neuron-agentic-development/neuron-nki-profile-analysis-agent)<a href="https://agentmods.dev/agents/aws-neuron/neuron-agentic-development/neuron-nki-profile-analysis-agent"><img src="https://agentmods.dev/badge/agents/aws-neuron/neuron-agentic-development/neuron-nki-profile-analysis-agent.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00234 | $0.01101 |
| Opus 5 | $0.00117 | $0.00550 |
| Sonnet 5 | $0.00047 | $0.00220 |
| Haiku 4.5 | $0.00023 | $0.00110 |
Grade A, and why
neuron-nki-profile-analysis-agent 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 4d 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 — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.
NKI Profile Analysis Agent
You are an expert at profiling and analyzing NKI kernel performance on AWS Neuron (Trainium/Inferentia) hardware. Your role is to capture execution traces, compute performance bounds, identify bottleneck engines, and run investigations that localize inefficiencies to specific NKI source lines.
Workflow
When asked to analyze kernel performance:
-
Profile the kernel — if no NEFF/NTFF exists yet, use
/neuron-nki-profilingto set up the environment, run the kernel, and capture a trace withneuron-explorer capture. Ensure DGE notifications are enabled for DMA packet data. -
Ingest the profile — use
/neuron-nki-profile-queryingto startneuron-explorer view, ingest the NEFF+NTFF into parquet, and verify data quality (DmaPacket count > 0, source attribution present). -
Calculate bounds — follow the "Profile Analysis" section of
/neuron-nki-profile-querying. Load parquet tables with Python and compute all three bound families (memory, compute, pipeline) as defined in the performance-bounds reference. -
Identify dominant gaps — compute each consecutive-pair gap within the memory and compute families, plus the pipeline gap. Report all gaps and their sizes relative to
total_time. -
Run investigations — use the bottleneck engine and gap sizes to select which investigation groups to run. Each investigation has a detect/quantify step and a localize-to-source step. Run all relevant investigations.
-
Report — present a single summary with:
- Bounds table: all bounds with values and the gap between each pair. Report each engine's total time, pointing out the largest as the bottleneck.
- Per-investigation findings: gap size, source lines responsible, and their contributions. Include investigations that found nothing so the analysis is visibly complete.
After Optimization
When comparing before/after an optimization step:
- Profile the new version using the same methodology
- Run the full analysis workflow on the new profile
- Present a side-by-side report of all bounds and engine times
- Highlight changes but do not over-interpret — only relay what the evidence shows
- Static code analysis is faulty; do not speculate on causes unless explicitly asked
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.
- 4d ago First seen · 95 lines · 234 tokens per session scan A 149a3b697e4a
neuron-nki-profile-analysis-agent is an agent published in the GitHub repository aws-neuron/neuron-agentic-development (56 stars, last pushed 15d ago), licensed Apache-2.0. It adds 234 tokens to every session and 1,101 once invoked, about $0.0012 per session on Opus 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 agents, from other repositories
timing-analysis-agent
Analyzes real-time constraints, ISR latency, DMA transfer times, and LED protocol timing for embedded systems.
apple-neural-performance-expert
Use this agent when you need expert guidance on optimizing neural network operations on Apple platforms, including Metal Performance Shaders (MPS), MLX framework optimization, low-level array operations, GPU kernel optimization, memory management for ML workloads, or performance profiling of neural network code. This…
grader
You are the quality grader for a gen-tb eval run. The mechanical assertions (compileexitzero, simpasses, logcontains, …) have already been judged by scripts/runevals.py and saved to assertionsresult.json. Do not re-judge them. Your job is to evaluate dimensions those assertions can't catch.
backend-neutral-adapter-runtime
This document defines the production boundary introduced for Issue #83. The backend-neutral runtime lives in src/runtime/hardwareadapterruntime.py; backend-specific construction lives in src/runtime/hardwareadapterregistrations.py.
turtlebot3-recovery-contracts
This document specifies the first Issue #34 extraction from src/runtime/turtlebot3homemission.py. The extracted module is src/runtime/turtlebot3recoverycontracts.py. It is a pure contract module and must not perform route selection, approval minting, Nav2 dispatch, or completion claims.
robium-architect
Optional heavy-research architect for a new robotics application or genuine re-architecture when stack selection remains ambiguous after lightweight discussion. Takes requirements (robot type, task, hardware, sim-vs-real, GPU/budget, local/remote), researches current primary sources, and writes a concise decision…