batch-process

Processes multiple content pieces through a sequential, checkpointed 10-phase pipeline with priority scheduling.

1|Updated Aug 18, 2026
One-click install
npx skills add https://github.com/teachskillofskills-ai/ContentForge-techshu --skill batch-process-teachskillofskills-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: batch-process
Source: https://github.com/teachskillofskills-ai/ContentForge-techshu/tree/main/skills/batch-process
Command: npx skills add https://github.com/teachskillofskills-ai/ContentForge-techshu --skill batch-process-teachskillofskills-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Producing dozens of articles one at a time is slow and error-prone, and interrupted runs lose all progress. This Skill turns a queue of content requirements into a hands-off batch run where each piece goes through the full ContentForge pipeline with all quality gates, and any interruption can be resumed from checkpoints. ## Core Features & Use Cases - Multi-source intake: Load requirements from local JSON, Google Sheets, Airtable, or CSV with per-row validation and priority-based queue sorting. - Checkpointed sequential execution: Each piece runs the full 10-phase pipeline with all 10 quality gates; transient failures auto-retry from checkpoints and persistent failures escalate to human review without stopping the queue. - Event-driven progress and reporting: A live status table updates after each piece or phase event, and a completion report summarizes APPROVED, review_required, and failed pieces with quality scores and output locations. - Use Case: An agency needs 15 blog posts for 3 clients by end of week — point the Skill at a Google Sheet of requirements and receive 14 approved .docx files plus one flagged for review, with a batch summary report. ## Quick Start Run /contentforge:batch-process with the URL of your requirements spreadsheet to produce every queued piece through the full pipeline.

Frequently Asked Questions about batch-process

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I batch produce multiple blog posts with AI?

Run /contentforge:batch-process with a requirements source such as a Google Sheet, CSV, Airtable base, or local JSON queue. Each row is validated, sorted by priority, and processed through the full 10-phase pipeline with all quality gates, producing one .docx per piece.

What input formats does batch content processing support?

Intake supports local JSON (the default tracking backend), Google Sheets, Airtable, and CSV files. Each row needs a requirement ID, content type, title, target audience, brand, word count target, priority from 1 to 5, and status.

Can I resume an interrupted batch content run?

Yes. Batch state lives in the tracking backend plus per-piece checkpoint directories on disk. Re-running the command skips completed rows and resumes the in-flight piece from its last gate-passed phase.

Does batch processing run content pieces in parallel?

No. Execution is strictly sequential, one pipeline at a time, because shared per-brand state, API rate limits, and context limits make in-session parallelism unsafe. Throughput comes from checkpointed resume rather than concurrency.

What happens when a piece fails quality review in a batch?

A piece scoring 5.0-6.9 after loop limits is marked review_required, never silently completed. Persistent failures such as missing brand profiles or two consecutive pipeline failures are escalated for human review while the queue continues.