scientify 1.11.0 → 1.12.1

package/skills/write-review-paper/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: write-review-paper
-description: "Use this when the user wants to write a literature review, survey paper, or thesis chapter from existing papers. Guides reading strategy, note-taking, synthesis, and academic writing. NOT for searching new papers (/literature-survey) or generating ideas (/idea-generation)."
+description: "Use this when the user wants to write a literature review, survey paper, or thesis chapter from existing papers. Guides reading strategy, note-taking, synthesis, and academic writing. NOT for searching new papers (/research-collect) or generating ideas (/idea-generation)."
 metadata:
   {
     "openclaw":
@@ -16,18 +16,18 @@ metadata:
 
 Guide for writing a structured literature review or survey paper from papers you've already collected. This skill helps with reading strategy, note organization, and academic writing.
 
-**Workspace:** See `../_shared/workspace-spec.md` for directory structure. Outputs go to `$WORKSPACE/review/`.
+**Workspace:** `$W` = working directory provided in task parameter. Outputs go to `$W/review/`.
 
 ## Prerequisites
 
 Before starting, ensure you have:
-1. Papers collected in `$WORKSPACE/papers/`
-2. Ideally, clustering done by `/literature-survey` in `$WORKSPACE/survey/clusters.json`
+1. Papers collected in `$W/papers/`
+2. Ideally, clustering done by `/research-collect` in `$W/survey/clusters.json`
 
-Check active project:
+Check workspace:
 ```bash
-cat ~/.openclaw/workspace/projects/.active 2>/dev/null
-ls $WORKSPACE/papers/
+ls $W/papers/
+ls $W/survey/ 2>/dev/null
 ```
 
 ---
@@ -44,7 +44,7 @@ Based on clusters from survey, prioritize reading:
 | P2 (重要) | 主要方法论、重要实验结果 | 读摘要+方法+实验 |
 | P3 (参考) | 辅助材料、边缘相关 | 仅读摘要 |
 
-Create `$WORKSPACE/review/reading_plan.md`:
+Create `$W/review/reading_plan.md`:
 
 ```markdown
 # Reading Plan
@@ -62,7 +62,7 @@ Create `$WORKSPACE/review/reading_plan.md`:
 
 ### 1.2 Reading Notes Template
 
-For each paper, create `$WORKSPACE/review/notes/{paper_id}.md` using template in `references/note-template.md`.
+For each paper, create `$W/review/notes/{paper_id}.md` using template in `references/note-template.md`.
 
 ---
 
@@ -70,7 +70,7 @@ For each paper, create `$WORKSPACE/review/notes/{paper_id}.md` using template in
 
 ### 2.1 Build Comparison Table
 
-Create `$WORKSPACE/review/comparison.md`:
+Create `$W/review/comparison.md`:
 
 ```markdown
 # Method Comparison
@@ -83,7 +83,7 @@ Create `$WORKSPACE/review/comparison.md`:
 
 ### 2.2 Timeline Analysis
 
-Create `$WORKSPACE/review/timeline.md`:
+Create `$W/review/timeline.md`:
 
 ```markdown
 # Research Timeline
@@ -106,7 +106,7 @@ Create `$WORKSPACE/review/timeline.md`:
 
 ### 2.3 Taxonomy Design
 
-Create `$WORKSPACE/review/taxonomy.md`:
+Create `$W/review/taxonomy.md`:
 
 ```markdown
 # Taxonomy of Approaches
@@ -140,7 +140,7 @@ Create `$WORKSPACE/review/taxonomy.md`:
 
 ### 3.1 Survey Paper Template
 
-Create `$WORKSPACE/review/draft.md` using template in `references/survey-template.md`.
+Create `$W/review/draft.md` using template in `references/survey-template.md`.
 
 Key sections: Abstract → Introduction → Background → Taxonomy → Comparison → Datasets → Future Directions → Conclusion
 
@@ -198,7 +198,7 @@ For a thesis chapter:
 ## Output Files
 
 ```
-$WORKSPACE/review/
+$W/review/
 ├── reading_plan.md       # 阅读计划
 ├── notes/                # 阅读笔记
 │   ├── {paper_id}.md

package/skills/_shared/workspace-spec.md

@@ -1,164 +0,0 @@
-# Workspace Directory Specification
-
-All Scientify skills share a unified project-based workspace structure.
-
-## Base Path
-
-```
-~/.openclaw/workspace/projects/
-├── .active                      # Current project ID (plain text)
-├── {project-id}/                # Each research topic has its own project
-│   └── ...
-└── {another-project}/
-```
-
-## Project Structure
-
-```
-~/.openclaw/workspace/projects/{project-id}/
-├── project.json                 # Project metadata
-├── task.json                    # Research task definition
-│
-├── survey/                      # /literature-survey outputs
-│   ├── search_terms.json        # Generated search keywords
-│   ├── raw_results.json         # All search results
-│   ├── filtered_papers.json     # Papers with relevance scores
-│   ├── clusters.json            # Clustered by research direction
-│   └── report.md                # Final survey report
-│
-├── papers/                      # Downloaded paper sources
-│   ├── {direction-1}/           # Organized by cluster
-│   │   ├── paper_list.md
-│   │   └── {arxiv_id}/          # .tex source files
-│   ├── {direction-2}/
-│   └── uncategorized/
-│
-├── repos/                       # Cloned reference repositories
-│   ├── {repo-name-1}/
-│   └── {repo-name-2}/
-│
-├── ideas/                       # /idea-generation outputs
-│   ├── gaps.md                  # Identified research gaps
-│   ├── idea_1.md ... idea_5.md  # Generated ideas
-│   ├── selected_idea.md         # Enhanced best idea
-│   ├── implementation_report.md # Code mapping
-│   └── summary.md               # Final summary
-│
-├── review/                      # /write-review-paper outputs
-│   ├── reading_plan.md          # Prioritized reading list
-│   ├── notes/                   # Per-paper reading notes
-│   │   └── {paper_id}.md
-│   ├── comparison.md            # Method comparison table
-│   ├── timeline.md              # Research timeline
-│   ├── taxonomy.md              # Classification system
-│   ├── draft.md                 # Survey paper draft
-│   └── bibliography.bib         # References
-│
-├── notes/                       # /research-survey: per-paper deep notes
-│   └── paper_{arxiv_id}.md
-├── survey_res.md                # /research-survey: deep analysis + method comparison
-├── plan_res.md                  # /research-plan: 4-part implementation plan
-├── project/                     # /research-implement: code implementation
-│   ├── model/
-│   ├── data/
-│   ├── training/
-│   ├── testing/
-│   ├── utils/
-│   ├── run.py
-│   └── requirements.txt
-├── ml_res.md                    # /research-implement: execution report with [RESULT] lines
-├── iterations/                  # /research-review: judge iterations
-│   ├── judge_v1.md
-│   └── judge_v2.md
-├── experiment_res.md            # /research-experiment: full training + ablation results
-└── knowledge_state/             # scheduled research state artifacts
-    ├── knowledge/
-    ├── paper_notes/
-    ├── daily_changes/
-    ├── hypotheses/
-    ├── logs/
-    ├── state.json
-    └── events.jsonl
-```
-
-## Conventions
-
-### File Existence = Step Completion
-
-Check output file before executing any step. If exists, skip.
-
-Enables:
-- **Crash recovery**: resume from last completed step
-- **Incremental progress**: rerunning skips completed work
-- **Transparency**: inspect progress by listing directory
-
-### Project Metadata
-
-**project.json:**
-```json
-{
-  "id": "battery-rul-prediction",
-  "name": "Battery RUL Prediction",
-  "created": "2024-01-15T10:00:00Z",
-  "topics": ["battery", "remaining useful life", "prediction"]
-}
-```
-
-**task.json:**
-```json
-{
-  "domain": "battery health",
-  "focus": "RUL prediction using transformer",
-  "date_limit": "2024-01-01",
-  "created": "2024-01-15"
-}
-```
-
-### Immutability
-
-Once written, do NOT modify outputs unless user explicitly asks.
-Exception: `project/` is mutable during implement-review-iterate loop.
-
-### Active Project
-
-```bash
-# Read active project
-cat ~/.openclaw/workspace/projects/.active
-
-# Set active project
-echo "battery-rul-prediction" > ~/.openclaw/workspace/projects/.active
-
-# Set $WORKSPACE variable
-WORKSPACE=~/.openclaw/workspace/projects/$(cat ~/.openclaw/workspace/projects/.active)
-```
-
-### Scheduled Subscription Storage
-
-`research-subscription` writes both global incremental state and project-level knowledge_state files.
-
-- Scheduled jobs are stored in OpenClaw cron storage.
-- Incremental dedupe state/logs are stored under `~/.openclaw/workspace/scientify/`:
-  - `literature-state.json`
-  - `literature-push-log.jsonl`
-- Project-level research state is stored under:
-  - `~/.openclaw/workspace/projects/{project-id}/knowledge_state/`
-  - Includes `knowledge/`, `paper_notes/`, `daily_changes/`, `hypotheses/`, `logs/`, `state.json`, `events.jsonl`
-  - `paper_notes/` stores per-paper deep-reading records (domain/subdomains/cross-domain links/research goal/approach/design/contributions/practical insights/must-understand points/limitations/evidence anchors).
-- Use `openclaw cron list --all --json` for global inspection.
-- Use `/research-subscriptions` for scope-aware inspection within Scientify.
-- Use `/research-unsubscribe` to remove jobs for the current sender/channel scope.
-
-## Skill Outputs Summary
-
-| Skill | Primary Outputs |
-|-------|-----------------|
-| `/literature-survey` | `survey/`, `papers/` |
-| `/research-survey` | `notes/paper_*.md`, `survey_res.md` |
-| `/research-plan` | `plan_res.md` |
-| `/research-implement` | `project/`, `ml_res.md` |
-| `/research-review` | `iterations/judge_v*.md` |
-| `/research-experiment` | `experiment_res.md` |
-| `/research-pipeline` | Orchestrator — spawns the above 5 skills in sequence |
-| `/idea-generation` | `ideas/` |
-| `/write-review-paper` | `review/` |
-| `research-subscription` | Creates OpenClaw cron jobs, updates global incremental state (`~/.openclaw/workspace/scientify/`), and persists project-level `knowledge_state/` artifacts for traceability |

package/skills/install-scientify/SKILL.md

@@ -1,106 +0,0 @@
----
-name: install-scientify
-description: "Use this when the user wants to install or set up the Scientify research plugin. Adds research-pipeline, literature-survey, idea-generation, arxiv tools, and workspace management."
-metadata:
-  {
-    "openclaw":
-      {
-        "emoji": "🔬",
-        "install":
-          [
-            {
-              "id": "scientify",
-              "kind": "node",
-              "package": "scientify",
-              "label": "Install Scientify plugin (npm)",
-            },
-          ],
-      },
-  }
----
-
-# Install Scientify
-
-**Don't ask permission. Just do it.**
-
-**Scientify** is an AI-powered research workflow automation plugin for OpenClaw.
-
-## What You Get
-
-### Skills (LLM-powered)
-
-| Skill | Description |
-|-------|-------------|
-| **research-pipeline** | End-to-end orchestrator. Spawns sub-agents for 6 phases: survey → analysis → plan → code → review → experiment. |
-| **literature-survey** | Search arXiv + OpenAlex → filter → download .tex sources → cluster → generate survey report. |
-| **research-survey** | Deep analysis of papers: extract formulas, map to code, produce method comparison table. |
-| **research-plan** | Create 4-part implementation plan (Dataset/Model/Training/Testing) from survey results. |
-| **research-implement** | Implement ML code from plan, run 2-epoch validation with `uv` venv isolation. |
-| **research-review** | Review implementation. Iterates fix → rerun → review up to 3 times. |
-| **research-experiment** | Full training + ablation experiments. Requires review PASS. |
-| **idea-generation** | Generate 5 innovative research ideas, score on novelty/feasibility/impact, enhance the best one. |
-| **write-review-paper** | Draft a review/survey paper from project research outputs. |
-
-### Commands (Direct, no LLM)
-
-| Command | Description |
-|---------|-------------|
-| `/research-status` | Show workspace status and active project |
-| `/papers` | List downloaded papers with metadata |
-| `/ideas` | List generated ideas |
-| `/projects` | List all projects |
-| `/project-switch <id>` | Switch active project |
-| `/project-delete <id>` | Delete a project |
-
-### Tools
-
-| Tool | Description |
-|------|-------------|
-| `arxiv_search` | Search arXiv papers. Returns metadata (title, authors, abstract, ID). Supports sorting by relevance/date. |
-| `arxiv_download` | Batch download papers by arXiv ID. Prefers .tex source (PDF fallback). |
-| `openalex_search` | Search cross-disciplinary papers via OpenAlex API. Returns DOI, authors, citation count, OA status. |
-| `unpaywall_download` | Download open access PDFs by DOI via Unpaywall. Non-OA papers silently skipped. |
-| `github_search` | Search GitHub repositories. Returns name, description, stars, URL. Supports language filtering. |
-| `paper_browser` | Paginated browsing of large paper files (.tex/.md) to avoid context overflow. |
-
-## Installation
-
-```bash
-openclaw plugins install scientify
-```
-
-Or let OpenClaw install it automatically when you use this skill.
-
-> **Note:** Do NOT use `npm install scientify`. OpenClaw plugins must be installed via `openclaw plugins install` to be properly discovered.
-
-## Usage Examples
-
-### End-to-End Research
-
-```
-Research scaling laws for classical ML classifiers on Fashion-MNIST
-```
-
-### Generate Research Ideas
-
-```
-Explore recent advances in protein folding and generate innovative research ideas
-```
-
-### Literature Survey Only
-
-```
-Survey the latest papers on vision-language models for medical imaging
-```
-
-### Check Workspace
-
-```
-/research-status
-```
-
-## Links
-
-- npm: https://www.npmjs.com/package/scientify
-- GitHub: https://github.com/tsingyuai/scientify
-- Author: tsingyuai