# Content Refinement Agent

Step 5 of the PaperOrchestra pipeline (arXiv:2604.05018).

- Repository: https://github.com/Ar9av/PaperOrchestra (694 stars)
- Category: Engineering
- License: see repository
- Last updated: 2026-10-11
- Runs scripts: yes, review before use

## Install

Claude Code: `npx skills add Ar9av/PaperOrchestra --skill content-refinement-agent -g -a claude-code -y`
Codex: `npx skills add Ar9av/PaperOrchestra --skill content-refinement-agent -g -a codex -y`
Cursor: `npx skills add Ar9av/PaperOrchestra --skill content-refinement-agent -g -a cursor -y`
Gemini CLI: `npx skills add Ar9av/PaperOrchestra --skill content-refinement-agent -g -a gemini-cli -y`

## Description

Step 5 of the PaperOrchestra pipeline (arXiv:2604.05018). Iteratively refine drafts/paper.tex by simulating peer review and applying targeted revisions, with strict accept/revert halt rules, deterministic 0-100 decision bands (Accept/Minor/Major/Reject) that drive a target-met early stop, and a Devil's Advocate concession-threshold guard that blocks acceptance on unresolved critical findings. Maintains a worklog and snapshots each iteration so revert is real, not symbolic. TRIGGER when the orchestrator delegates Step 5 or when the user asks to "refine the draft", "iterate on the paper", or "run peer review on this paper".

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

# Content Refinement Agent (Step 5)

Faithful implementation of the Content Refinement Agent from PaperOrchestra
(Song et al., 2026, arXiv:2604.05018, §4 Step 5, App. F.1 pp. 49, 51).

**Cost: ~5, 7 LLM calls** (App. B), typically ~3 refinement iterations, each
consisting of one reviewer call and one revision call.

The paper highlights this step as one of the largest contributors to overall
quality: refinement alone accounts for +19% (CVPR) and +22% (ICLR) absolute
acceptance-rate improvement (Fig. 4). Get this step right.

## Inputs

- `workspace/drafts/paper.tex`, output of Step 4
- `workspace/inputs/conference_guidelines.md`
- `workspace/inputs/experimental_log.md`, used as ground truth for the
  hallucination check
- `workspace/citation_pool.json` / `workspace/refs.bib`, the allowed
  bibliography

## Outputs

- `workspace/refinement/iter1/`, `iter2/`, `iter3/`, per-iteration snapshots
  containing `paper.tex`, `paper.pdf`, `review.json`, `score.json`
- `workspace/refinement/worklog.json`, append-only history of decisions
- `workspace/final/paper.tex` and `workspace/final/paper.pdf`, copy of the
  best accepted snapshot

## The refinement loop

```
prev_score = score(paper.tex)                  # baseline from initial draft
snapshot iter0/

for iter in 1..ITER_CAP (default 3):
    1. simulate_review(paper.tex) → review.json
       (uses `references/reviewer-rubric.md` rubric)

    2. apply_revision(paper.tex, review.json) → new_paper.tex
       (uses verbatim Refinement Agent prompt at `references/prompt.md`)

    3. snapshot iter<N>/ with new_paper.tex, review.json
       latexmk -pdf new_paper.tex → iter<N>/paper.pdf

    4. score(new_paper.tex) → curr_score

    5. decide via score_delta.py:
       - if curr.overall > prev.overall:                       ACCEPT
       - elif curr.overall == prev.overall and net_subaxis ≥0: ACCEPT
       - else:                                                 REVERT

    6. apply_worklog.py to append the decision

The rest: https://github.com/Ar9av/PaperOrchestra/blob/HEAD/skills/content-refinement-agent/SKILL.md