{"version":"2.1.224","anchor":"skills-can-be-loaded-from-a-mounted-github-repository","canonical_anchor":"skills-can-be-loaded-from-a-mounted-github-repository","heading":"Skills can be loaded from a mounted GitHub repository","tier":"notice","area":"Skills","scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.224\/e\/skills-can-be-loaded-from-a-mounted-github-repository","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.224","markdown":"### Skills can be loaded from a mounted GitHub repository\n\nSkills can load from a mounted repo's .claude\/skills, in cloud sandboxes only.\n\n**What**\n\nA new docs section describes loading skills from a mounted repository's root `.claude\/skills` directory. Skills are discovered once per session from the code as it stood when the session started, and this works in cloud sandboxes only, not self-hosted.\n\n**Details**\n\n- Carries an explicit warning that repository skills are agent instructions running inside your trust boundary, so anyone who can commit to the repo can change agent behaviour.\n\n**Evidence**\n\n`Skills from a GitHub repository`\n\n- Area: Skills\n- Names: `.claude\/skills`\n- Tier: You'll notice\n- Useful: 3\/5\n- Signal: 3\/5"}