Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
 
← Back to list

Content Refinement Agent (PaperOrchestra Step 5)

Iteratively improves a LaTeX paper draft via simulated peer review, with deterministic accept/revert decisions and real snapshot rollbacks.

Content & WritingAdvanced★ 638⑂ 90AI score 8/10Last updated: Aug 9, 2026

What it does

  • Runs a simulated peer review of workspace/drafts/paper.tex using an AgentReview-style 6-axis rubric, emitting axis scores, an overall score, and a decision band (Accept ≥80 / Minor 65–79 / Major 50–64 / Reject <50).
  • Applies targeted revisions and snapshots every iteration (iter1/, iter2/… with paper.tex, paper.pdf, review.json, score.json) so reverts are real, not symbolic.
  • Decides ACCEPT / REVERT / plateau early-stop / target-met stop deterministically through score_delta.py exit codes.
  • Adds a Devil's Advocate reviewer plus concession_guard.py that blocks acceptance while a CRITICAL finding stands.
  • Enforces iron rules: no new experiments, every number must trace to experimental_log.md, no defensive limitation-listing to game the scorer.

Who it's for

  • Researchers and grad students writing conference papers in LaTeX who want an automated pre-submission review loop.
  • Users already running the PaperOrchestra pipeline (Steps 1–5).
  • Anyone who wants auditable score trajectories and rollback history for a draft.

Example uses

  1. "Run peer review on this paper" → get strengths, weaknesses, questions, 6-axis scores and a decision band.
  2. "Iterate on the paper three times" → snapshots iter1–iter3 and promotes the best accepted version to workspace/final/paper.tex.
  3. If iteration 2 regresses, it auto-reverts to iter1, halts, and reports a trajectory like "iter0 58.0 Major → iter1 67.3 Minor (accept) → iter2 revert".

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/Ar9av/PaperOrchestra/HEAD/skills/content-refinement-agent/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the skills/content-refinement-agent folder from the GitHub repo Ar9av/PaperOrchestra into my ~/.claude/skills/content-refinement-agent/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/Ar9av/PaperOrchestra.git && mkdir -p ~/.claude/skills && cp -r PaperOrchestra/skills/content-refinement-agent ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal in the folder where you keep tools.
  2. Clone the repo: git clone https://github.com/Ar9av/PaperOrchestra.git
  3. Copy the skill: mkdir -p ~/.claude/skills && cp -r PaperOrchestra/skills/content-refinement-agent ~/.claude/skills/
  4. It references skills/shared/ and skills/paper-orchestra/scripts/, so copy those too: cp -r PaperOrchestra/skills/shared PaperOrchestra/skills/paper-orchestra ~/.claude/skills/
  5. Install prerequisites: a LaTeX toolchain with latexmk (TeX Live / MacTeX) and Python 3.
  6. Prepare your workspace files: workspace/drafts/paper.tex, workspace/inputs/conference_guidelines.md, workspace/inputs/experimental_log.md, and your bibliography.
  7. Restart Claude Code, then say "refine the draft" or "run peer review on this paper" to trigger the skill.
View source on GitHub ↗License: NOASSERTION