Clo-Author¶
external · status: active · focus: end-to-end · discipline: economics · started: 2026
Project page: https://github.com/hugosantanna/clo-author
Source: projects/landscape/clo-author.yml
Positioning¶
A Claude Code scaffold for empirical economics research, spanning literature review through journal submission. Eighteen agents organized as worker-critic pairs (e.g., Strategist + strategist- critic, Coder + coder-critic) with thirteen named pipeline commands from /discover and /strategize through /write, /review, /revise, /talk, and /submit. Targets the full inputs → knowledge production → outputs arc of the RISE diagram with a domain-specific focus on applied economics.
Distinctive contribution¶
Worker-critic pair architecture as a first-class design principle: every creator agent has a paired critic that cannot edit files, with 3-round escalation when they disagree. Built-in peer-review simulation with intellectual-disposition reviewers (Structuralist, Credibility, Measurement, Policy, Theory, Skeptic) weighted by journal culture and FATAL / ADDRESSABLE / TASTE classification of comments. Direct sibling to e2er but distributed as a forkable Claude Code template rather than a Docker stack.
Evaluation scores¶
| Dimension | Score (0–3) | Note |
|---|---|---|
| Lifecycle coverage | 3 | Thirteen stages including dissemination (/submit) — broadest coverage in catalog after E2ER. |
| Autonomy level | 2 | Supervised: researcher approves plans before execution; worker-critic disagreements escalate to human after 3 rounds. |
| Architectural transparency | 3 | Skill, agent, and command definitions are plain markdown in .claude/; full guide at hugosantanna.github.io/clo-author/. |
| Inputs supported | 2 | Topic / question / referee-report inputs; user provides data; multi-language analysis (R, Python, Julia). |
| Outputs / reproducibility | 2 | Generates replication packages with /submit; LaTeX paper is single source of truth; quality gate (>=80/100) before ship. |
| Internal evaluation | 2 | Built-in peer-review simulation with weighted reviewer dispositions; no published external benchmark yet. |
| Openness | 1 | Source is public on GitHub but no declared open-source license — reuse rights are uncertain. |
| Maturity / traction | 1 | Created 2026-02; 193 stars; very active development through May 2026; single-author project at this date. |
| Cross-family policy | 0 | Worker-critic pairs within Claude Code (single model family). |
| Runtime assurance | 2 | Worker-critic pair architecture with 3-round escalation + 80/100 quality gate before ship. |
| Cross-platform portability | 1 | Claude Code primary back-end; multi-language analysis (R, Python, Julia) is data-side, not agent-side. |
Scored on 2026-05-18. See the evaluation rubric.
Tags¶
Pipeline stages: rq-formulation hypothesis-generation literature-discovery literature-synthesis research-design data-acquisition data-analysis formal-modeling code-generation paper-drafting revision-editing referee-simulation dissemination
Architectural features: multi-agent debate-consensus human-in-loop tool-use persistent-memory artifact-versioning
Inputs: research-topic research-question referee-report
Outputs: paper-latex code slides replication-package referee-reports
Data sources: user-provided
Knowledge sources: literature
Limitations¶
- No declared open-source license; reuse rights uncertain.
- Economics-focused by default; portability to other fields requires customizing domain-profile + journal profiles.
- Depends on Claude Code; not a fully local solution.
- Quality gates and review simulation are internal; no third-party validation reported.
Related projects in this catalog¶
Related references (literature catalog)¶
- Wu, J. et al. (2025). Agentic Reasoning: A Streamlined Framework for Enhancing LLM Reasoning with Agentic Tools
wu2025agenticreasoning cunningham2025claudecode(BibTeX)eberhardt2025claudecode(BibTeX)