Skip to content

Security: ORBWEVA/academy

Security

SECURITY.md

Security Policy

Reporting a vulnerability

Email security@orbweva.com with subject [academy] <short summary>. Please do not open a public GitHub issue.

We aim to acknowledge within 72 hours.

What this installer does (threat model)

@orbweva/academy performs a narrow set of operations on your machine:

  1. Shells out to git to shallow-clone public GitHub repos under https://github.com/ORBWEVA/* into a temp directory.
  2. Reads files from the cloned temp dirs and copies skills/<name>/ folders into either ~/.claude/skills/ (global) or ./.claude/skills/ (project-local).
  3. Prints suggested shell commands for CLI tool installs (brew, winget, scoop, npm -g) and MCP server setup (claude mcp add …). These are not executed automatically — you must copy/paste them.
  4. Removes the temp directory after copying.

It does not: write to ~/.claude.json, edit shell dotfiles, run arbitrary commands from any repo, reach any server besides GitHub, or transmit telemetry.

Trust boundaries

  • Source of skills — only repos listed in manifest.json under the ORBWEVA/* namespace. A malicious PR attempting to add a third-party repo must pass review.
  • Runtime dependencies — none (beyond Node 18+ and system git). Zero node_modules.
  • npx trust — running npx @orbweva/academy executes arbitrary npm-published code. Audit the published version by running npm view @orbweva/academy before installing, or git clone this repo and run node bin/install.js directly.

Disclosure timeline

We follow coordinated disclosure:

  1. Reporter emails security@orbweva.com.
  2. We triage within 72 hours.
  3. Fix developed privately; release candidate shared with reporter.
  4. Patch released + CVE filed if applicable.
  5. Public advisory 7–30 days after patch release, depending on severity.

There aren’t any published security advisories