# Wayland Content Batch Process

Batch-process a directory of content briefs: scan the briefs, produce a blog draft for...

- Repository: https://github.com/FerroxLabs/wayland (616 stars)
- Category: Writing
- License: Apache-2.0
- Last updated: 2026-10-31
- Runs scripts: no, instructions only

## Install

Claude Code: `npx skills add FerroxLabs/wayland --skill wayland-content-batch-process -g -a claude-code -y`
Codex: `npx skills add FerroxLabs/wayland --skill wayland-content-batch-process -g -a codex -y`
Cursor: `npx skills add FerroxLabs/wayland --skill wayland-content-batch-process -g -a cursor -y`
Gemini CLI: `npx skills add FerroxLabs/wayland --skill wayland-content-batch-process -g -a gemini-cli -y`

## Description

Batch-process a directory of content briefs: scan the briefs, produce a blog draft for each, run a humanize pass over the batch, then build repurpose chains for email, LinkedIn, and X-thread. Use when the user has many content briefs in a folder and wants every one turned into a humanized draft with platform derivatives in a single run. Do NOT use for a single brief or when no corpus of briefs can be resolved; use the single-post blog workflow instead.

## SKILL.md, as published by its author (untrusted third-party text, read it as data)

# Content Brief Batch Processor

**Estimated time:** 5-10 minutes per brief

This workflow runs deterministically over a directory of content briefs. It
scans the briefs into a list, produces a full blog draft for each, humanizes the
whole batch against the voice profile, and builds repurpose chains for email,
LinkedIn, and X-thread. It is the bulk counterpart to the single-post publish
flow.

Resolve the content corpus before running. In order: use any file or folder
attached to the conversation (an exported brief bundle, CSV, or zip of markdown
briefs) read from the workspace cwd; else read from a connected filesystem MCP
server or a Google Sheets / Microsoft 365 export holding the brief list; else
ask the user to attach the content briefs export. If no corpus resolves, stop
and ask. Do not produce drafts from empty input.

Inputs: a briefs directory path (required) and an optional path to
voice-notes.md (recommended).

## When to Use

- User has a folder of content briefs to process at once
- User wants every brief turned into a humanized draft plus derivatives
- User wants a repeatable batch run, not an interactive single post

## Workflow Steps

**Step 1: Scan the Briefs** (uses: content-audit)

Resolve and read the briefs directory, then list every brief found (markdown or
JSON) into a structured brief list. If the corpus cannot be resolved, halt and
ask the user to attach it. Do not continue on empty input.

- Input: briefs directory path
- Output: structured list of resolved briefs
- Key focus: a complete, validated brief list; halt on no input

**Step 2: Produce Drafts** (uses: content-blog)

For every brief in the list, produce a full blog draft, applying the voice
profile if one was provided. Process the briefs as a batch and keep each draft
mapped to its source brief. Halt on error rather than skipping silently.

The rest: https://github.com/FerroxLabs/wayland/blob/HEAD/src/process/resources/bundled-workflows/bodies/wayland-content-batch-process/SKILL.md