Maintainers: Bryce Wang (Stanford REAP / CoPaper.AI)
Compatibility:claude-codecodex
Despite the "Awesome" name this is not a link list: the repo ships 4,150 first-party SKILL.md skills across 299 installable Claude-plugin packs — per-journal depth packs (~12 lifecycle skills each: topic selection, identification, robustness, tables/figures, writing style, referee strategy, rebuttal, replication package, submission) for top venues in economics, finance, management, accounting, and beyond, plus discipline-wide breadth bundles (one venue-fit skill per journal, 100+ journals each) and a cross-journal Research-Toolkit. Content is genuinely journal-specific (editors, acceptance norms, house style, canon), not find-replace boilerplate. This manifest catalogs 25 representative skills: the full 7-skill cross-journal Research-Toolkit, the full 12-skill QJE depth pack as the exemplar, the workflow routers of five further depth packs (Econometrica, JPE, REStud, Management Science, MIS Quarterly), and one breadth-bundle venue-fit skill (AER). The AER depth pack lives in the separate brycewang-stanford/AER-skills repo, cataloged in RISE under the auto-empirical-research-skills pack.
Plans and prioritizes the robustness suite and online appendix QJE referees expect, exploiting the journal's no-page-limit norm, without changing the core identification design.
Runs and audits the empirical analysis rather than just advising — maps a DiD/IV/RDD/synthetic-control/DML design to the concrete StatsPAI / Stata MCP tools in the environment and reports the fitted, audited number.
Stress-tests the causal identification strategy (RCT, staggered DID, IV, RDD, event study) against the QJE general-interest bar — including the journal-specific credibility ranking and the "novel dataset answering a first-order question" exception — before tables are drafted.
Polishes prose, abstract, and introduction so the big idea lands fast for a general-interest reader, reflecting QJE's house style of long, ambitious, narrative-driven empirical papers.
Finalizes main exhibits for a figure-forward QJE manuscript — clean tables and self-contained notes that read well in QJE's single-PDF, author-date format.
Router for the 12-skill Econometrica depth pack (ecta-*) — topic selection through rebuttal, with the pack's theory-model and identification skills tuned to Econometrica's formal bar.
Router for the 12-skill Management Science depth pack (mgsci-*) — topic selection through rebuttal, including picking the right INFORMS Department lane (analytical vs empirical).
Router for the 12-skill MIS Quarterly depth pack (misq-*) — topic selection through revision, first naming which of the four IS research traditions (behavioral, design science, economics, organizational) the paper lives in.
Router for the QJE pack — sequences manuscript work from topic selection through rebuttal for a Quarterly Journal of Economics submission; routes, does not replace, the specialized qje-* skills.
Router for the 12-skill Review of Economic Studies depth pack (restud-*) — topic selection through rebuttal, weighing design-based and structural identification on REStud's equal terms.
Sequences a manuscript across venues — which cross-journal toolkit skill to invoke next, from picking a target journal through execution, submission-readiness, simulated review, rebuttal, and the replication package.
Stakes the manuscript's contribution against the frontier — what it builds on, what it overturns, and why it belongs in a general-interest journal; positions the contribution rather than writing a full survey.
Adds the model or conceptual framework that turns a credible empirical coefficient into a lesson with broad implications — the frame QJE expects around a headline result.
Assembles and validates the Data-Editor replication package against the target venue's data-and-code policy — master script, pinned environment, README/roadmap, restricted-data plan, script-to-exhibit output map, and a Data-Editor-grade checklist.
War-games likely referee objections before submitting, to close gaps pre-emptively given QJE's ~2-week desk screen and ~1-4% acceptance rate; anticipates reports rather than writing the post-decision response.
Pre-submission peer-review rehearsal — a calibrated Associate Editor desk-screen plus 2-3 distinct-lens referees for the target venue, adversarially verified and synthesized into a report, decision band, and prioritized skill-mapped fix list.
Structures the revision and response letter when a QJE decision letter arrives (R&R or reject-and-resubmit); routes re-analysis to the relevant qje-* skill rather than redoing it.
Turns a referee report into a point-by-point response letter plus revision plan — editor first, every comment answered, each empirical fix backed by a real re-run via the target pack's skill and the execution bridge.
Venue-fit skill for AER from the English Social-Science Journal bundle — fit, framing, the method-and-evidence bar, house style, official-submission re-check, and desk-reject heuristics; exemplar of the one-skill-per-journal breadth-bundle format.
Final pre-submission preflight for QJE via Editorial Express — single-PDF format, double-blind anonymization, author-date references, the no-fee policy, and supplementary files.
Answers "which journal should I send this to?" — profiles the paper, shortlists candidate venues across 185+ packs, and ranks them into reach/match/safe with a resubmission ladder, reading live venue facts from each pack's source-map.