Instruction file imported from sx-tane/context-os (.github/instructions/customization.instructions.md). Copyright stays with the author.

Customization Authoring Instructions

Name: customization
Author: sx-tane

Skill

For the full authoring workflow, skeletons, wiring rules, and completion checklist, apply the contextos-authoring skill.

Customization Shape

Skills live at .github/skills/<name>/SKILL.md and .codex/skills/<name>/SKILL.md mirrors; keep assets/ and references/ subdirectories aligned with the procedure.
Add scripts/ only when the skill has an executable helper; do not create empty script placeholders.
Agents wire skills with a skill reference, skeleton link, checklist link, and inline fallback rules.
Instruction files stay short and point to the paired skill for deep implementation detail.

Documentation Checklist

Update .codex/README.md, AGENTS.md, and the relevant .github/agents/README.md or .github/instructions/README.md when agent, instruction, skill, routing, or folder structure behavior changes. Do not add a top-level .github/README.md only to satisfy benchmarks.
Update the nearest folder README when the customization changes how files in that folder should be edited.
Keep links to skeletons, checklists, examples, and scripts valid after every rename or move.

Do Not

Do not duplicate long skill procedures inside instruction files.
Do not leave skill reference tables stale after adding or renaming a skill.
Do not add unused script placeholders when no executable helper exists.

{ "ocm": "1", "id": "sx-tane-context-os-customization-instructions", "kind": "skill", "name": "customization", "description": "Use when creating or modifying ContextOS agents, instruction files, or skills. Keeps customization routing, folder structure, and README documentation aligned.", "publisher": "sx-tane", "version": "1.0.0", "capabilities": { "domains": [ "general" ], "tags": [ "skill-md", "github-instructions" ], "languages": [ "en" ] }, "quality_prior": 0.6, "examples": [ "Use when creating or modifying ContextOS agents, instruction files, or skills. Keeps customization routing, folder structure, and README documentation aligned." ], "primary": false, "metadata": { "source": { "provider": "github-instructions", "repository": "https://github.com/sx-tane/context-os", "path": ".github/instructions/customization.instructions.md", "ref": "3f1449f1995131be68353670f6bfb0b2ee1552b0", "url": "https://github.com/sx-tane/context-os/blob/3f1449f1995131be68353670f6bfb0b2ee1552b0/.github/instructions/customization.instructions.md", "key": "sx-tane/context-os/.github/instructions/customization.instructions.md" }, "applies_to": ".github/{agents,instructions,skills}/**/*.md,.codex/{agents,instructions,skills}/**/*.md" }, "instructions": "# Customization Authoring Instructions\n\n## Skill\n\nFor the full authoring workflow, skeletons, wiring rules, and completion checklist, apply the **contextos-authoring** skill.\n\n## Customization Shape\n\n- Skills live at `.github/skills/<name>/SKILL.md` and `.codex/skills/<name>/SKILL.md` mirrors; keep `assets/` and `references/` subdirectories aligned with the procedure.\n- Add `scripts/` only when the skill has an executable helper; do not create empty script placeholders.\n- Agents wire skills with a skill reference, skeleton link, checklist link, and inline fallback rules.\n- Instruction files st", "cost": { "context_tokens": 352 } }

customization

About

Customization Authoring Instructions

Skill

Customization Shape

Documentation Checklist

Do Not

Use it

Manifest

Reviews