Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →ar9av/content-refinement-agent
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 orche
Passed 14 of 19 test cases with the skill, and 5 of 19 without it.
and 9 more in the full report
12 fixed7 passed either way6 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 12 test cases the skill fixed. Same model, same prompt, with and without the skill.
We generated workspace/reverse_outline.md using reverse_outline.py during Gate C. Which reviewer rubric axis consumes workspace/reverse_outline.md during the simulated peer review call?
| Checked | Without skill | With skill |
|---|---|---|
| Passes workspace/reverse_outline.md into the reviewer call specifically for scoring the Logical Flow axis. | Missed | Passed |
The instructions your model reads when the skill loads.
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.
workspace/drafts/paper.tex — output of Step 4workspace/inputs/conference_guidelines.mdworkspace/inputs/experimental_log.md — used as ground truth for the
hallucination checkworkspace/citation_pool.json / workspace/refs.bib — the allowed
bibliographyworkspace/refinement/iter1/, iter2/, iter3/ — per-iteration snapshots
containing paper.tex, paper.pdf, review.json, score.jsonworkspace/refinement/worklog.json — append-only history of decisionsworkspace/final/paper.tex and workspace/final/paper.pdf — copy of the
best accepted snapshotprev_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
test cases passed, vs 5 / 19 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.