What is profiler-mcp CLAUDE.md?

Project instructions for profiler-mcp, a package that connects AI assistants to Intel VTune and AMD uProf performance profilers. It also describes mock command-line tools for testing without profiler hardware.

Use it when developing, testing, or changing profiler-mcp, especially its VTune and uProf command handling, mock tools, fixtures, return data, and self-tests.

It gives the assistant the project’s commands, architecture rules, test expectations, and fixed data contracts so changes do not quietly break either profiler server.

How Stankye/profiler-mcp configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to Stankye/profiler-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/Stankye/profiler-mcp/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/Stankye/profiler-mcp

Made for: Claude Code.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/stankye/profiler-mcp/claude-md.svg)](https://agentmods.dev/instructions/stankye/profiler-mcp/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/stankye/profiler-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/stankye/profiler-mcp/claude-md.svg" alt="Measured on agentmods" height="20"></a>

Per session 654 This file is loaded in full into every session.

When invoked 654 The same file — it is already loaded in full.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00654	$0.00654
Opus 5	$0.00327	$0.00327
Sonnet 5	$0.00131	$0.00131
Haiku 4.5	$0.00065	$0.00065

Measured 6d ago against content hash 9f23be79287d, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

profiler-mcp CLAUDE.md 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 6d 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.

CLAUDE.md · 47 lines

How it starts

The opening of the file, as written. The whole thing — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.

profiler-mcp

Two MCP servers — vtune (Intel VTune) and uprof (AMD uProf) — one package, one core, fully self-testable without profiler hardware via byte-realistic mock CLIs.

Commands

uv sync                      # deps (Python >=3.11, mcp SDK v1.x pinned <2)
uv run pytest                # deterministic gate: ALL of layers 1-4. Keep green.
uv run ruff check src tests
make -C examples/workloads   # compile the profiling workload (needed by e2e tests)
uv run vtune-mcp             # run a server on stdio (uprof-mcp likewise)

Load-bearing invariants

tests/fixtures/* are verbatim vendor output (provenance: docs/research/captures-*.md). Never edit them to make code pass — fix the code. New fixtures require new verbatim captures in docs/research/ first.
Mock CLIs (src/profiler_mcp/mocks/) are strict: unknown flags are errors, so server command-construction bugs surface as test failures. If you add a flag in vtune/cli.py/uprof/cli.py, teach the mock the same flag in the same change.
All tool return shapes live in core/models.py (TypedDicts = wire contract). Change shapes there only, and expect both servers plus tests to move together.
Every result is stamped with its mode (real/mock); anything user-visible that reports numbers must carry the mock caveat. Never let synthetic numbers pass as real.
The two servers stay structurally parallel: a capability added to one gets added to the other (or the asymmetry documented, like uprof_timechart).
mcp stays pinned >=1.28,<2 until the v2 migration is done deliberately.

Architecture in one breath

server.py (tool surface, FastMCP) → cli.py (argv construction) → core/runner.py (subprocess) → real binary or mock (resolved by core/config.py; mode auto-detects) → output → parse.py → TypedDicts from core/models.py; results registered in core/store.py under short ids (hs-0000); comparisons via core/compare.py return a verdict that fail_on_regression turns into a hard ToolError (the agentic-loop gate).

Read the full file on GitHub · 47 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.

6d ago First seen · 47 lines · 654 tokens per session scan A 9f23be79287d

Subscribe to this mod's changes

profiler-mcp CLAUDE.md is an instructions file published in the GitHub repository Stankye/profiler-mcp (0 stars, last pushed 1mo ago), licensed MIT. It adds 654 tokens to every session, about $0.0033 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 2026-08-31.

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,182 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

profiler-mcp: Instructions file for Claude Code