paper-orchestration

Orchestrate multi-stage research paper drafting with planning gates and review checkpoints.

Updated May 15, 2026
One-click install
npx skills add https://github.com/lth0/codexSkill --skill paper-orchestration
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: paper-orchestration
Source: https://github.com/lth0/codexSkill/tree/main/skills/agents/research-writing/paper-orchestration
Command: npx skills add https://github.com/lth0/codexSkill --skill paper-orchestration

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

paper-orchestration prevents low-quality “one prompt writes the paper” failures by forcing the controller to detect the current stage, require evidence-driven inputs, and gate drafting through plans, task packets, and reviews.

Core Features & Use Cases

  • Stage-detected writing workflow: Automatically classifies work into scope, evidence, method, experiments, drafting, and review stages, then routes to the appropriate next skills.
  • Plan-first hard gates: Blocks manuscript text until required planning artifacts exist (project overview, outline, progress, task packet, and—when applicable—chapter architecture).
  • Full-paper redraft safety: Locks chapter architecture before drafting and forbids inheriting an outdated chapter split when venue or paper type changes.
  • Multi-agent chapter ownership: Enforces one fresh agent per major chapter with disjoint edit ownership, records provenance in plan/chapter-agent-provenance.md, and requires a controller review gate.
  • Auditability and quality checks: Requires persistent task packets, two review gates (spec compliance + quality), and a capability-use audit written to plan/progress.md.

Quick Start

Use paper-orchestration when your research-writing task needs a structured, multi-stage controller workflow for a medium or full-paper revision with repeated quality failures.

Frequently Asked Questions about paper-orchestration

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

FAQPage Schema
How do I prevent uncontrolled one-pass manuscript generation when drafting research papers?

Research paper writing can be controlled by applying a multi-agent orchestration workflow with staged gates. This approach blocks manuscript text generation until planning artifacts exist, requiring evidence-driven inputs and structured revision cycles to prevent low-quality drafts.

What is stage-detected routing in a multi-agent research writing workflow?

Stage-detected routing automatically classifies writing tasks into scope, evidence, method, experiments, drafting, and review stages. It then routes each stage to the appropriate next skills to coordinate evidence and consistent peer review workflows.

How do I manage full-paper redrafting safely when the venue or paper type changes?

Full-paper redrafting requires locking the chapter architecture before drafting begins. This prevents inheriting an outdated chapter split and enforces plan-first hard gates to ensure the new manuscript structure matches the updated venue requirements.

Can I use multi-agent chapter ownership to prevent edit conflicts during paper drafting?

Multi-agent chapter ownership assigns one fresh agent per major chapter with disjoint edit ownership. This prevents overlapping edits and records provenance in a dedicated file, requiring a controller review gate before finalizing sections.

Does structured paper drafting require spec compliance and quality review gates?

Structured paper drafting requires two review gates: spec compliance and quality checks. Completion is gated by these reviews plus a capability-use audit written to a persistent progress file to ensure auditability.

When should I not use a staged orchestration workflow for research writing?

Staged orchestration workflows are not suited for small text edits or unstructured drafting tasks. They are designed for medium-sized revisions and full-paper drafting where coordinating evidence, methods, and experiments requires strict planning gates.