Claude Code Plugins

Community-maintained marketplace

Feedback

research-pipeline-runner

@WILLOSCAR/research-units-pipeline
1
0

|

Install Skill

1Download skill
2Enable skills in Claude

Open claude.ai/settings/capabilities and find the "Skills" section

3Upload to Claude

Click "Upload skill" and select the downloaded ZIP file

Note: Please verify skill by going through its instructions before using it.

SKILL.md

name research-pipeline-runner
description Run this repo’s Units+Checkpoints research pipelines end-to-end (survey/综述/review/调研/教程/系统综述/审稿), with workspaces + checkpoints. **Trigger**: run pipeline, kickoff, 继续执行, 自动跑, 写一篇, survey/综述/review/调研/教程/系统综述/审稿. **Use when**: 用户希望端到端跑流程（创建 `workspaces/<name>/`、生成/执行 `UNITS.csv`、遇到 HUMAN checkpoint 停下等待）。 **Skip if**: 用户明确要手工逐条执行（用 `unit-executor`），或你不应自动推进到 prose 阶段。 **Network**: depends on selected pipeline (arXiv/PDF/citation verification may need network; offline import supported where available). **Guardrail**: 必须尊重 checkpoints（无 Approve 不写 prose）；遇到 HUMAN 单元必须停下等待；禁止在 repo root 创建 workspace 工件。

Research Pipeline Runner

Use this skill to run the repo’s UNITS + checkpoints workflow end-to-end, while keeping semantic work LLM-driven (scripts are helpers, not the author).

Non-negotiables

  • Use UNITS.csv as the execution contract; one unit at a time.
  • Respect checkpoints (CHECKPOINTS.md): no long prose until required approvals are recorded in DECISIONS.md (survey default: C2).
  • Stop at HUMAN checkpoints and wait for explicit sign-off.
  • Never create workspace artifacts (STATUS.md, UNITS.csv, DECISIONS.md, etc.) in the repo root; always use workspaces/<name>/.
  • Treat skill scripts as helpers (deterministic tasks / scaffolding). For semantic units, the “real work” is done by following the relevant SKILL.md and refining outputs until quality gates pass.

Decision tree: pick a pipeline

User goal → choose:

  • Survey/综述/调研 + Markdown draft → pipelines/arxiv-survey.pipeline.md
  • Survey/综述/调研 + PDF output → pipelines/arxiv-survey-latex.pipeline.md
  • Snapshot/速览 → pipelines/lit-snapshot.pipeline.md
  • Tutorial/教程 → pipelines/tutorial.pipeline.md
  • Systematic review/系统综述 → pipelines/systematic-review.pipeline.md
  • Peer review/审稿 → pipelines/peer-review.pipeline.md

Recommended run loop (human-like)

  1. Create a workspace under workspaces/ and kick off the pipeline.
  2. Run units in --strict mode so “scaffolds that look finished” don’t silently pass.
  3. Treat each blocked unit as a mini writing loop:
    • read the unit’s SKILL.md
    • refine the artifacts until they look like real work (not templates)
    • then mark the unit DONE and continue
  4. Stop once for the human checkpoint (survey default: C2), then proceed to writing after approval.

Practical commands (optional helpers)

  • Kickoff + run (recommended): python scripts/pipeline.py kickoff --topic "<topic>" --pipeline <pipeline-name> --run --strict
  • Resume: python scripts/pipeline.py run --workspace <ws> --strict
  • Approve checkpoint: python scripts/pipeline.py approve --workspace <ws> --checkpoint C2
  • Mark refined unit: python scripts/pipeline.py mark --workspace <ws> --unit-id <U###> --status DONE --note "LLM refined"

Handling common blocks

  • HUMAN approval required: summarize produced artifacts, ask for approval, then record it and resume.
  • Quality gate blocked (output/QUALITY_GATE.md exists): treat current outputs as scaffolding; refine per the unit’s SKILL.md; mark DONE; resume.
  • No network: ask the user for an export; use arxiv-search offline import.
  • Weak coverage (mapping/notes): broaden queries or reduce/merge subsections before writing.

Quality checklist

  • UNITS.csv statuses reflect actual outputs (no DONE without outputs).
  • No prose is written unless DECISIONS.md explicitly approves it for the relevant checkpoint/sections.
  • The run stops at HUMAN checkpoints with clear next questions.
  • In --strict mode, scaffold/stub outputs do not get marked DONE without LLM refinement.