Skip to content

Security: Doryski/appgraph

Security

SECURITY.md

Security Policy

Supported Versions

Security fixes are applied to the latest released minor version. As the project is pre-1.0, only the most recent 0.x release is supported.

Version Supported
0.1.x ✅
< 0.1 ❌

Reporting a Vulnerability

Please do not report security vulnerabilities through public GitHub issues.

Instead, report them privately via GitHub Security Advisories. This lets us discuss and fix the issue before it is publicly disclosed.

When reporting, please include:

  • A description of the vulnerability and its potential impact
  • Steps to reproduce or a proof of concept
  • The affected version(s)
  • Any suggested mitigation, if known

You can expect an initial acknowledgement within a few days. Once the issue is confirmed and fixed, a patched release will be published and the advisory disclosed with appropriate credit, if desired.

Threat model

appgraph is a static analysis tool. It reads source files with the TypeScript parser and writes YAML and HTML into an output directory. It is useful to know what that does and does not involve:

  • It never executes your application code. Analysis is parser-only: source files are turned into ASTs via ts.createSourceFile and read. Your components, loaders and route modules are not imported or evaluated.
  • It does execute your appgraph.config.*. A .ts/.mts/.cts config is transpiled to a temporary .mjs file next to the config file (removed afterwards) and loaded with a dynamic import(); a .js/.mjs config is imported directly. Either way it runs with the full privileges of the invoking user, like any other JavaScript config file. This applies to analyze() too, which loads the config itself when you pass no config object. Treat a config file from an untrusted source the same way you would treat an untrusted vite.config.ts.
  • It refuses to write through symlinks. The CLI will not write an artifact whose path, or whose parent directory, is a symlink, so a planted link in the output directory cannot redirect a write elsewhere. Files are staged to a temporary sibling and renamed into place, and nothing is replaced until every file has been staged.
  • Text from your repository is sanitised before it reaches your terminal. Control characters (C0 except newline and tab, DEL and C1) in diagnostic codes, messages, file names and traces are printed as \xNN escapes, so a crafted file name or string constant cannot inject escape sequences into the CLI output.
  • It makes no network requests and sends no telemetry.
  • The generated HTML report embeds strings taken from your source — route paths, component names, guard expressions, endpoint URLs, test-id values. Those values are escaped when rendered, and the YAML artifacts quote and escape any scalar containing control characters, line separators or other non-printable code points (then re-parse the result as a self-check), but the output is still a derivative of your code: if your routes or guard expressions contain sensitive information, so will the report. Review the output directory before publishing or committing it.
  • The generated artifacts describe your app's structure, including which screens require authentication and which sit behind feature flags. This is reconnaissance material. Do not publish it for a private application.

If you believe any of the above statements is false for a given input, that is a security report — please use the advisory link above.

Thank you for helping keep the project and its users safe.

There aren't any published security advisories