torchtalk-setup

torchtalk-setup is a skill for Claude Code from opendatahub-io/ai-helpers. It costs 19 tokens per session (2,568 once invoked), scanned B, original, Apache-2.0.

A setup guide for TorchTalk, a server that lets coding agents inspect how PyTorch connects Python, C++, and CUDA code.

In plain words
What is it for?
It helps install TorchTalk, obtain PyTorch source code, register the server, and verify the setup.
Why use it?
It prepares the required software and configuration before cross-language PyTorch analysis can work.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: reads .claude/ paths; mentions Claude Code.

Part of the odh-pytorch plugin — 3 skills shipped together

Good fit It helps install TorchTalk, obtain PyTorch source code, register the server, and verify the setup.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/opendatahub-io/ai-helpers/torchtalk-setup
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.

Any agent
npx skills add opendatahub-io/ai-helpers --skill torchtalk-setup
Clone the repo
git clone --depth 1 https://github.com/opendatahub-io/ai-helpers

Made for: Claude Code.

Or install odh-pytorch, the plugin that ships this one along with the rest of its 3 skills.

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-setup

README.md
[![agentmods](https://agentmods.dev/badge/skills/opendatahub-io/ai-helpers/torchtalk-setup/github.svg)](https://agentmods.dev/skills/opendatahub-io/ai-helpers/torchtalk-setup)
Your own site
<a href="https://agentmods.dev/skills/opendatahub-io/ai-helpers/torchtalk-setup"><img src="https://agentmods.dev/badge/skills/opendatahub-io/ai-helpers/torchtalk-setup/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.

agentmods 80×15 button for torchtalk-setup

Your own site · 80×15
<a href="https://agentmods.dev/skills/opendatahub-io/ai-helpers/torchtalk-setup"><img src="https://agentmods.dev/badge/skills/opendatahub-io/ai-helpers/torchtalk-setup.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 19 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,568 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 7 findings, 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 Privilege Escalation · line 44
    Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.
    Fix: Avoid sudo/root unless strictly required. Prefer least-privilege patterns. If elevation is needed, document the justification and scope.
  • medium Privilege Escalation · line 45
    Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.
    Fix: Avoid sudo/root unless strictly required. Prefer least-privilege patterns. If elevation is needed, document the justification and scope.
  • medium Privilege Escalation · line 120
    Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.
    Fix: Avoid sudo/root unless strictly required. Prefer least-privilege patterns. If elevation is needed, document the justification and scope.
  • medium Privilege Escalation · line 121
    Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.
    Fix: Avoid sudo/root unless strictly required. Prefer least-privilege patterns. If elevation is needed, document the justification and scope.
  • medium Privilege Escalation · line 364
    Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.
    Fix: Avoid sudo/root unless strictly required. Prefer least-privilege patterns. If elevation is needed, document the justification and scope.
  • medium Rogue Agent · line 88
    Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.
    Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
  • medium Privilege Escalation · line 363
    Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.
    Fix: Avoid sudo/root unless strictly required. Prefer least-privilege patterns. If elevation is needed, document the justification and scope.
How audits are shown
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 5.1 $0.00019 $0.02568
Opus 5 $0.00010 $0.01284
Sonnet 5 $0.00004 $0.00514
Haiku 4.5 $0.00002 $0.00257

Measured 9d ago against content hash bbe1663f36cc, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade B, and why

torchtalk-setup scanned grade B with 1 finding 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 9d 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.

Asks for rootmediumPrivilege escalation

A mod that escalates privileges can change anything on the machine, not only the project.

- Ubuntu/Debian: sudo apt install python3.12
plugins/odh-pytorch/skills/torchtalk-setup/SKILL.md · 418 lines

How it starts

The opening of the file, as written. The whole thing — 418 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Name

odh-ai-helpers:torchtalk-setup

Synopsis

/torchtalk:setup

Description

The torchtalk:setup command guides you through the complete installation and configuration of the TorchTalk MCP server, which enables cross-language binding analysis for PyTorch codebases (Python/C++/CUDA). This command:

  • Checks for and installs required dependencies (Python 3.10+, pip)
  • Clones and builds TorchTalk from source
  • Locates or clones PyTorch source code
  • Configures TorchTalk with the PyTorch source path
  • Registers the MCP server with Claude Code at user scope
  • Verifies the complete setup

After setup, you can use /torchtalk:trace and the full suite of TorchTalk MCP tools across all Claude Code sessions.

Implementation

Phase 1: Check Prerequisites

1.1 Verify Python Version
python3 --version

Minimum required: Python 3.10

If Python is too old or missing:

Python 3.10+ is required for TorchTalk.

Install or upgrade Python:
- macOS: brew install [email protected]
- Ubuntu/Debian: sudo apt install python3.12
- Fedora/RHEL: sudo dnf install python3.12
1.2 Verify pip
python3 -m pip --version

If pip is missing, provide platform-specific installation instructions.

1.3 Check for Existing TorchTalk Installation
which torchtalk

If torchtalk is found:

  • Run torchtalk status to show current configuration
  • Ask the user if they want to:
    1. Reconfigure the existing installation
    2. Reinstall from scratch
    3. Exit setup

Phase 2: Install TorchTalk

2.1 Determine Installation Location

Ask the user where they want to install TorchTalk:

Where would you like to install TorchTalk?

1. ~/src/torchtalk (recommended)
2. ~/.local/share/torchtalk
3. Custom path

Store the chosen directory as $TORCHTALK_DIR.

2.2 Clone the Repository
mkdir -p "$(dirname "$TORCHTALK_DIR")"
git clone https://github.com/adabeyta/torchtalk.git "$TORCHTALK_DIR"

Handle errors:

  • Directory already exists: Ask to use existing or remove and re-clone
  • Network issues: Provide troubleshooting steps
  • Permission issues: Suggest a different directory

Read the full file on GitHub · 418 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. 9d ago First seen · 418 lines · 19 tokens per session scan B bbe1663f36cc

Subscribe to this mod's changes

torchtalk-setup is a skill published in the GitHub repository opendatahub-io/ai-helpers (37 stars, last pushed 5d ago), licensed Apache-2.0. It adds 19 tokens to every session and 2,568 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.

Related

Other skills, from other repositories

manimce-best-practices

Trigger when: (1) User mentions "manim" or "Manim Community" or "ManimCE", (2) Code contains from manim import , (3) User runs manim CLI commands, (4) Working with Scene, MathTex, Create(), or ManimCE-specific classes. Best practices for Manim Community Edition - the community-maintained Python animation engine.…

calesthio/OpenMontage · 149 tokens

ampl-callbacks

Expert in AMPL solver callbacks via ampls (AMPL Solver Libraries): amplpygurobi, amplpycplex, amplpyxpress, amplpycopt. Use for GenericCallback, lazy constraints, user cuts, subtour elimination, Benders cuts added to the solver matrix, heuristic solution injection, custom stopping criteria, and solve monitoring. Knows…

marcos-dv/ampl-agents · 105 tokens

ampl-tuner

Expert in AMPL/MP solver options across all supported solvers. Recommends and configures lim:time, mip:gap, threads, IIS, barrier, and solver-specific knobs. Fetches official option pages from dev.ampl.com before citing names. Use when tuning solve performance, setting optimality gaps, choosing LP/MIP/NLP algorithms…

marcos-dv/ampl-agents · 99 tokens

ampl-modeler

Expert AMPL modeler for LP, MIP, MINLP, NLP, CP, and quadratic optimization. Use PROACTIVELY when the user needs a new model, reformulation, multi-objective design, logical constraints, supply-chain or energy formulations, nonlinear modeling. Knows MP modeling guide, Colab/MO-Book examples, and dev.ampl.com solver…

marcos-dv/ampl-agents · 91 tokens

ampl-energy

Expert AMPL modeler for electric power and energy systems — unit commitment, optimal power flow (DC/AC/BIM), capacity expansion, hydrothermal scheduling, BESS, dispatch, and energy market formulations. Uses readable logical constraints and MP solvers. Use for power grid models, MINLP unit commitment, conic OPF, or…

marcos-dv/ampl-agents · 89 tokens

ampl-python

Expert in amplpy and Python integration with AMPL. Use when writing or refactoring amplpy scripts, notebooks, FastAPI services, data pipelines (pandas/polars), solver configuration via Python, result extraction, or Colab/MO-Book Python workflows. Knows amplpy.ampl.com and dev.ampl.com amplpy best practices.

marcos-dv/ampl-agents · 75 tokens