mlview-local
Safety scan
A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.
Badge for your README
Markdown
[](/listings/realmyang--mlview)HTML
<a href="/listings/realmyang--mlview"><img src="/api/listings/realmyang--mlview/badge.svg" alt="findsafeskills safety rating" /></a>Install
Install instructions from the README
As written by the repo's author. findsafeskills has not verified these steps.
Each release attaches the extension (a `.vsix` file) and the two skill ZIPs to its page under Releases.
- Download the `.vsix` file. Install it with VS Code's **Extensions: Install from VSIX...** command. - Download the skill ZIP for your assistant and extract it at the root of your project. [The skill, per assistant](docs/user-guide.md#the-skill-per-assistant) says which ZIP is for which assistant. - For Claude Code, the skill is also packaged as a [plugin](claude-plugin/README.md).
To work from a clone of this repository instead, build the extension as [docs/development.md](docs/development.md#prerequisites) describes. You need Python 3.10 or newer with the packages of `requirements-dev.txt` in a virtual environment, and Node 20.18.1 or newer. Then copy the skill into your project with the installer:
python3 tools/install_skill.py /path/to/your/project
The installer copies the skill into the project's `.agents/skills/mlview/` folder. For Claude Code, add `--destination .claude/skills/mlview`.
Use one skill location in a project, not both. [The user guide](docs/user-guide.md) has the details.
What was scanned
The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.
.claude-plugin/marketplace.json · Manifest · 24 lines
View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)
- {
- "name": "mlview-local",
- "owner": { "name": "MLView" },
- "metadata": {
- "description": "MLView's marketplace: the plugin from this checkout, and the same plugin fetched from GitHub.",
- "version": "0.4.0"
- },
- "plugins": [
- {
- "name": "mlview",
- "source": "./claude-plugin",
- "description": "A skill for Claude Code: read an ML project, write a workflow diagram that cites its source lines, and check and publish it with a local helper."
- },
- {
- "name": "mlview-github",
- "source": {
- "source": "git-subdir",
- "url": "https://github.com/realmyang/MLView.git",
- "path": "claude-plugin"
- },
- "description": "The same MLView plugin, installed from the repository on GitHub."
- }
- ]
- }
README.md · README · 143 lines
View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)
- # MLView
- [](https://github.com/realmyang/MLView/actions/workflows/ci.yml)
- [](LICENSE)
- **Ask the assistant in your editor to explain an ML workflow, then explore its
- answer as an interactive diagram linked to the source.** MLView is a skill for
- GitHub Copilot, Codex and Claude Code, a document format, and a VS Code
- viewer. The assistant reads the code and configuration and writes the diagram:
- steps, connections and findings, with quotes from the files it read.
- **Preview.** The format and the viewer can still change between releases,
- and fixes go into the next release.
- 
- The diagram of `samples/configured_training`, with a step selected.
- ## How it works
- ```mermaid
- flowchart LR
- A[Your assistant + MLView skill] --> B[Read source and config]
- B --> C[LLM interpretation]
- C --> D[WorkflowDocument draft]
- D --> E[Local validation and publication]
- E --> F[Interactive VS Code diagram]
- F --> G[Source navigation]
- A -->|Explicit refinement| C
- ```
- - MLView needs no model API key of its own. Your assistant supplies the model,
- its tools and its permissions.
- - The skill's helper checks a draft's structure, cited paths and exact quotes,
- records file hashes, and publishes a `*.mlview.json` file.
- - Neither the helper nor the viewer imports or runs your code.
- - Opening, filtering and navigating a diagram never call a model.
- ## Install
- Each release attaches the extension (a `.vsix` file) and the two skill ZIPs
- to its page under Releases.
- - Download the `.vsix` file. Install it with VS Code's
- **Extensions: Install from VSIX...** command.
- - Download the skill ZIP for your assistant and extract it at the root of
- your project.
- [The skill, per assistant](docs/user-guide.md#the-skill-per-assistant)
- says which ZIP is for which assistant.
- - For Claude Code, the skill is also packaged as a
- [plugin](claude-plugin/README.md).
- To work from a clone of this repository instead, build the extension as
- [docs/development.md](docs/development.md#prerequisites) describes. You need
- Python 3.10 or newer with the packages of `requirements-dev.txt` in a
- virtual environment, and Node 20.18.1 or newer. Then copy the skill into
- your project with the installer:
- ```sh
- python3 tools/install_skill.py /path/to/your/project
- ```
- The installer copies the skill into the project's `.agents/skills/mlview/`
- folder. For Claude Code, add `--destination .claude/skills/mlview`.
- Use one skill location in a project, not both.
- [The user guide](docs/user-guide.md) has the details.
- ## A first diagram
- 1. Open your project in VS Code and start the `mlview` skill in your
- assistant. The [user guide](docs/user-guide.md#start-the-skill) says how.
- Ask, for example: "Explain training with train.py and configs/distill.json.
- Show where data enters, which model is updated, how the losses combine, how
- validation works, and what source inspection cannot establish. Tell me what
- looks wrong."
- 2. The assistant publishes a `*.mlview.json` file. Click it in the Explorer
- and it opens as the diagram. Select a step, connection or finding to read
- its claim and quotes. Press Enter to open the cited lines beside the
- diagram.
- 3. Ask the same assistant to refine the diagram. A valid new revision replaces
- the one on screen.
- ## Example
- [`samples/configured_training.mlview.json`](samples/configured_training.mlview.json)
- is a diagram of the small project in
- [`samples/configured_training/`](samples/configured_training/README.md). Open
- this repository in VS Code with the extension installed and click the file.
- The file records which assistant and model wrote it.
- ## What MLView checks, and what it does not
- - The helper and the extension check a document's structure, that every cited
- path is inside the workspace, the file hashes and the exact quotes.
- - A matching quote shows that the cited lines exist and have not changed. It
- does not show that they support the claim.
- - No automated check calls a model or opens VS Code.
- [What is checked, and what is not](docs/validation.md) says what the checks
- cover.
- - Treat a diagram as the assistant's reading of the code, and check it against
- the cited lines.
- ## In this repository
- | Path | What it holds |
- |---|---|
- | `skills/mlview/` | The skill: its instructions, reference files and helper, with the helper's tests |
- | `contracts/` | The JSON Schema of the format, and the cases the schema, the helper and the extension must agree on |
- | `vscode-extension/` | The VS Code extension |
- | `webview/` | The diagram viewer the extension shows |
- | `claude-plugin/` | The Claude Code plugin, with a generated copy of the skill |
- | `samples/` | Small example projects and example diagrams |
- | `tools/`, `scripts/` | Install, package, build and check tools, with their tests |
- | `docs/` | The user guide, the format reference, and pages on the architecture, development, releasing and what is checked |
- ## Develop
- Use a Python 3.10+ virtual environment and Node 20.18.1 or newer.
- ```sh
- python -m pip install -r requirements-dev.txt
- sh scripts/e2e.sh
- ```
- On Windows run `pwsh -File scripts/e2e.ps1`. The full check builds the
- viewer and the extension, type-checks and tests them, runs the Python tests
- and the documentation link check, and packages the two skill ZIPs and the
- VSIX. [CONTRIBUTING.md](CONTRIBUTING.md) has the steps for a change, and
- [docs/development.md](docs/development.md) has the commands for each part.
- ## More
- - The [user guide](docs/user-guide.md)
- - [The format and the helper's commands](docs/format.md), and the format's
- [JSON Schema](contracts/workflow.schema.json)
- - The [architecture](docs/architecture.md), [development](docs/development.md)
- and [what is checked, and what is not](docs/validation.md)
- - The skill: [skills/mlview/SKILL.md](skills/mlview/SKILL.md)
- - [Contributing](CONTRIBUTING.md), the [security policy](SECURITY.md) and the
- [code of conduct](CODE_OF_CONDUCT.md)
- - The [changelog](CHANGELOG.md), the [license](LICENSE) and the
- [third-party notices](THIRD_PARTY_NOTICES.md)
Ratings and comments
Sign in to rate this skill.
Sign inFree to use. If it’s helped, pass it on.