菜单
Use when running human-supervised iterative research experiments, recording autoresearch results, maintaining resume state, applying verify/guard gates, or coordinating local_mac smoke tests with remote_desktop_4060 formal runs.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use when coordinating a full research project, graduation thesis, dissertation, paper pipeline, literature-to-experiment-to-writing workflow, or when deciding which research, notebook, PDF, DOCX, citation, visualization, or audit skill to use next.
Use when planning, implementing, reviewing, or refactoring research experiment code, including experiment architecture, data pipelines, model or algorithm modules, training/evaluation scripts, config management, logging/output conventions, reproducibility checks, and mapping code runs to thesis experiments.
Use when performing final checks before submitting, defending, compiling, or revising a research paper, graduation thesis, dissertation, journal manuscript, conference paper, report, response letter, reference list, figure set, or defense presentation.
Use when planning a research paper, graduation thesis, dissertation, journal manuscript, conference paper, or project report before drafting; especially when the user has an advisor-given topic/title, needs Topic Intake to Research Blueprint, or needs a research question, contribution story, chapter outline, experiment-to-claim map, figure plan, or missing-materials checklist.
Use when building a literature review, related work section, citation matrix, paper search strategy, BibTeX candidate list, or citation plan for a research paper, thesis, dissertation, survey, journal article, or conference paper.
Use when planning, designing, reviewing, or generating research-paper figures, thesis figures, experiment plots, result tables, model diagrams, method pipelines, architecture schematics, Mermaid diagrams, captions, or publication-quality visualizations.
| name | research-autoresearch-loop |
| description | Use when running human-supervised iterative research experiments, recording autoresearch results, maintaining resume state, applying verify/guard gates, or coordinating local_mac smoke tests with remote_desktop_4060 formal runs. |
Use this skill for human-AI collaborative experiment iteration. It borrows useful ideas from autoresearch systems but keeps the user in control of research direction, claims, and final decisions.
docs/thesis/autoresearch-results.tsv.docs/thesis/autoresearch-state.json.docs/thesis/experiment-reports/EXP-*.md with baseline comparison, verify/guard status, and environment snapshot status.remote_desktop_4060; cloud_autodl is fallback.Read references/loop.md for the iteration contract. Read references/source-map.md for provenance.
local_mac smoke test.remote_desktop_4060 when needed.experiment-registry.md.scripts/new_autoresearch_iteration.py.scripts/new_experiment_report.py --experiment-id EXP-... --baseline EXP-... when a baseline comparison is relevant.$research-results-analysis.Always include: