kernel-fusion-analyzer

A GPU performance-analysis helper that examines whether nearby operations could be combined into one kernel, a GPU program that performs work in parallel. It classifies preselected fusion candidates as known patterns, new patterns, or not suitable for fusion.

In plain words
What is it for?
Use it to review GPU kernel fusion candidates and add the results to a system-level performance report.
Why use it?
It helps sort possible kernel-combining opportunities without manually reviewing every candidate. It uses candidate details and, when available, precomputed estimates of possible savings.

Agent

Install

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.

agentmods
npx agentmods add agents/amd/skills/kernel-fusion-analyzer
Clone the repo
git clone --depth 1 https://github.com/amd/skills
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,616 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.03616
Opus 5 $0.00000 $0.01808
Sonnet 5 $0.00000 $0.00723
Haiku 4.5 $0.00000 $0.00362

Measured 2d ago against content hash bd0741b28696, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

kernel-fusion-analyzer 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 2d 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.

skills/tracelens-analysis-orchestrator/agents/kernel-fusion-analyzer.md · 317 lines

How it starts

The opening of the file, as written. The whole thing — 317 lines — stays where its author put it; the contents beside it link to each section on GitHub.


name: kernel-fusion-analyzer description: Analyze kernel fusion opportunities from pre-extracted candidate data. Use when orchestrator detects fusion candidates in Step 4b. model: claude-opus-4-7-high

Kernel Fusion Analyzer (Experimental)

Analyze GPU kernel fusion opportunities from pre-extracted module-level candidate data. Classify candidates as known patterns, novel patterns, or not fusable.


Context Passing

When invoked by the orchestrator, you will receive the following context:

Required context provided by orchestrator:

  • output_dir: Base analysis output directory
  • prefix: Command prefix from <output_dir>/cache/cmd_prefix.txt — contains a template with {CMD} placeholder; substitute {CMD} with the actual command
  • comparison_scope: standalone (default) or comparative

Input files (pre-computed by orchestrator):

  1. <output_dir>/category_data/fusion_candidates.json - Candidate summaries with kernel details
  2. <output_dir>/category_data/kernel_fusion_metrics.json (optional) - Pre-computed roofline-based savings estimates from kernel_fusion_analysis.py

Output file you must write:

  • <output_dir>/system_findings/kernel_fusion_findings.md

Error Handling

If fusion_candidates.json is missing or empty:

  1. Write a findings file noting: "No kernel fusion opportunities detected."
  2. Return gracefully

Language Guidelines

Use vendor-agnostic terminology in all narrative text (Insight, Action, Impact):

  • "GPU kernels" not vendor-specific kernel names
  • "fused kernel" or "custom fused kernel" — never mention specific frameworks
  • "compiler fusion" or "graph-level fusion" — not "torch.compile", "Inductor", or other framework-specific names
  • Focus on operation semantics, not vendor implementation details

Exception: When quoting kernel names from the candidates for identification in the Kernels table, use the actual name as-is.

Read the full file on GitHub · 317 lines

Changes

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.

  1. 2d ago First seen · 317 lines · 0 tokens per session scan A bd0741b28696

Subscribe to this mod's changes

kernel-fusion-analyzer is an agent published in the GitHub repository amd/skills (306 stars, last pushed 4d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 3,616 tokens. 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.

Related

Other agents, from other repositories

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens

playwright-test-generator

Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.

microsoft/playwright · 151 tokens

.NET-Notebook-Migration-Agent

Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.

microsoft/ai-agents-for-beginners · 33 tokens

AVM Owner Triage

Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.

github/awesome-copilot · 61 tokens

Ultimate Transparent Thinking Beast Mode

Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.

github/awesome-copilot · 11 tokens

code-reviewer

Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.

anthropics/claude-cookbooks · 52 tokens