torchtalk CLAUDE.md

torchtalk CLAUDE.md is an instructions file for coding agents from TorchedHat/torchtalk. It costs 1,205 tokens per session, scanned A, original, MIT.

Project instructions for Torchtalk, an MCP server that analyzes how Python and C++ code connects inside PyTorch.

In plain words
What is it for?
They help inspect Python modules, C++ call graphs, and native bindings, then query the indexed PyTorch source through MCP tools.
Why use it?
They give developers the commands and architecture needed to install, test, lint, and run the analysis server correctly.

Instructions file

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 instructions/torchedhat/torchtalk/claude-md
Clone the repo
git clone --depth 1 https://github.com/TorchedHat/torchtalk

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.

agentmods badge for torchtalk CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/torchedhat/torchtalk/claude-md.svg)](https://agentmods.dev/instructions/torchedhat/torchtalk/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/torchedhat/torchtalk/claude-md"><img src="https://agentmods.dev/badge/instructions/torchedhat/torchtalk/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 1,205 This file is loaded in full into every session.
When invoked 1,205 The same file — it is already loaded in full.
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 $0.01205 $0.01205
Opus 5 $0.00602 $0.00602
Sonnet 5 $0.00241 $0.00241
Haiku 4.5 $0.00120 $0.00120

Measured 3d ago against content hash 10f357d69c3a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

torchtalk 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 3d 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 · 110 lines

How it starts

The opening of the file, as written. The whole thing — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

This file provides guidance to Claude Code when working with code in this repository.

Commands

# Install
pip install -e .
pip install -e ".[dev]"

# Test
pytest
PYTORCH_SOURCE=/path/to/pytorch pytest tests/test_binding_detector_pytorch.py

# Lint
ruff check src/torchtalk/
ruff format src/torchtalk/

# Run MCP server
python -m torchtalk mcp-serve --pytorch-source /path/to/pytorch

Project Structure

src/torchtalk/
├── server.py              # MCP server (7 tools: status + 6 query)
├── cli.py                 # CLI (torchtalk mcp-serve)
├── formatting.py          # Response formatting (CompactText/Markdown)
└── analysis/
    ├── binding_detector.py    # pybind11/TORCH_LIBRARY detection (tree-sitter)
    ├── cpp_call_graph.py      # C++ call graph extraction (libclang)
    ├── python_analyzer.py     # Python module/class analysis (AST)
    ├── patterns.py            # Search directories, exclusion patterns
    └── helpers.py             # Utility functions

Architecture

MCP server providing cross-language binding analysis for PyTorch.

Server (server.py): FastMCP-based, auto-builds and caches index from PyTorch source. Background thread for C++ call graph.

Analysis (analysis/): BindingDetector (tree-sitter), CppCallGraphExtractor (libclang, 60K+ functions), PythonAnalyzer (AST).

Data Sources: native_functions.yaml, derivatives.yaml, compile_commands.json, tree-sitter AST.

Cache: ~/.cache/torchtalk/ — bindings (~10MB), call graph (~50MB).

MCP Tools

IMPORTANT: Use mcp__torchtalk__* tools directly. Do NOT import or run Python code from torchtalk.server.

Tool Description
get_status() TorchTalk readiness summary across bindings, call graph, modules, tests
trace(func, focus?) Trace any PyTorch op: Python → YAML → C++ → file:line
search(query, mode?, backend?) mode="bindings": dispatch registrations. mode="kernels": CUDA kernel launches
graph(func, mode?, depth?, fuzzy_all_levels?, walk_python?, focus?) mode="callers": inbound. mode="calls": outbound. mode="impact": transitive callers (depth/fuzzy_all_levels/walk_python/focus apply to impact only)
modules(name, mode?, focus?) mode="trace": class details (focus="full" adds bases/docstring). mode="list": browse by category ("nn", "optim", "all")
tests(query?, mode?, limit?, focus?) mode="find": search tests (focus narrows to functions/classes/files). mode="utils": list utilities (query/focus ignored). mode="file_info": test file details
affected(funcs, depth?) Map changed C++ functions (comma-separated) to impacted Python test files

Read the full file on GitHub · 110 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. 3d ago First seen · 110 lines · 1,205 tokens per session scan A 10f357d69c3a

Subscribe to this mod's changes

torchtalk CLAUDE.md is an instructions file published in the GitHub repository TorchedHat/torchtalk (11 stars, last pushed 7d ago), licensed MIT. It adds 1,205 tokens to every session, about $0.0060 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.

Related

Other instructions, from other repositories

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

spec-kit AGENTS.md

AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.

github/spec-kit · 7,104 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

langchain AGENTS.md

AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.

langchain-ai/langchain · 4,345 tokens

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).

microsoft/vscode · 5,001 tokens

next.js 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