spatial-transcriptomics
Skill0 starsChanged 5 months agoMITView on GitHub
Orchestrator skill for spatial transcriptomics analysis. Routes to one of three personality modes (aristotle/plato/socrates) based on user intent. Covers platforms: Visium, Xenium, MERFISH, Slide-seq, Stereo-seq, CosMx, SpatialDB, Seq-Scope, DBiT-seq, STARmap. Frameworks: Scanpy, Squidpy, Giotto, Seurat, SpatialDE, BayesSpace, SPARK-X, SpaGCN, Cell2location, RCTD, Tangram. Trigger keywords: spatial transcriptomics, Visium, Xenium, MERFISH, Slide-seq, spatial gene expression, spatial omics, single-cell spatial, spatial analysis, spot deconvolution, spatial clustering, spatially variable genes, 空间转录组, 空间组学, 空间基因表达, 空间分析, 空间聚类, 空间差异表达, 10x Visium, Stereo-seq, CosMx, seqFISH, FISSEQ.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- zamushwani/biomedical-ai-skillstheir own version, first seen Aug 26, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- zamushwani/biomedical-ai-skills★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 46
Ambiguity resolution: When markers from multiple modes are present, use this priority: socrates > plato > aristotle.
- Distinctive: line 89
Usage rule: Always trigger BioMCP when the user asks about a specific gene, pathway, or disease context.
- Distinctive: line 106
Before starting any analysis, the skill should briefly scan the findings log for relevant past insights.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Spatial Transcriptomics — Mode Router
Decision Tree
Route user requests to one of three personality modes. Detect intent from keywords in the user's natural language question, then load the corresponding mode file.
User Question
│
├── Contains Socrates markers?
│ "全新" "颠覆" "换思路" "unconventional" "breakthrough"
│ "rethink" "from scratch" "无预设" "paradigm shift"
│ "abandon standard" "completely new direction"
│ → YES → Load modes/socrates.md
│
├── Contains Plato markers?
│ "新方法" "前沿" "cutting-edge" "recent" "improved"
│ "最新" "更好的方法" "state of the art" "alternative"
│ "compare methods" "beyond standard" "最新进展"
│ → YES → Load modes/plato.md
│
├── Contains Aristotle markers?
│ "标准流程" "常规分析" "standard pipeline" "routine analysis"
│ "best practice" "tried and tested" "gold standard"
│ "recommended workflow" "standard preprocessing" "QC"
│ → YES → Load modes/aristotle.md
│
└── No clear markers (DEFAULT)
→ Load modes/aristotle.md (safety-first)Ambiguity resolution: When markers from multiple modes are present, use this priority:
socrates > plato > aristotle. Novelty-seeking overrides safety defaults. If only
"novel" appears without stronger markers, route to plato mode.
Mode Overview
Aristotle (modes/aristotle.md)
Standard, reproducible spatial transcriptomics pipelines using well-established tools (Scanpy, Squidpy, Seurat). Follows community-vetted best practices, built-in QC thresholds, and published workflows. Aristotle was the empiricist who classified all of nature — this mode inherits his systematic rigor. Appropriate for routine analysis, benchmarking, and when reproducibility is paramount. No experimental methods are recommended.
Plato (modes/plato.md)
Standard pipelines augmented with systematic literature search via BioMCP. Searches PubMed for recent methods (last 2 years), compares alternatives, and recommends improvements over the baseline pipeline. Plato was the dialectician who explored every possibility — this mode inherits his breadth. Useful when the user wants to know about newer approaches without abandoning the standard framework entirely.
Socrates (modes/socrates.md)
Question-driven reverse design. Starts from the user's biological question, then designs a custom analysis strategy from scratch — tool selection, parameter choices, and interpretation framework are all derived from the question rather than from a predefined pipeline. Socrates was the gadfly who questioned everything until truth emerged — this mode inherits his relentless curiosity. No preset assumptions about which tools should be used.
BioMCP Integration
This skill uses the biomcp_biomcp MCP tool for biomedical queries. All modes may
invoke it, but Plato and Socrates modes rely on it heavily for literature-backed
method discovery.
| Command | Purpose | Used By |
search article -c "spatial transcriptomics" --date-from 2023 | Find recent methods | plato, socrates |
search article -c "<gene> spatial transcriptomics" | Gene-specific literature | all modes |
get gene <symbol> | Gene annotation, known roles | all modes |
search variant "<variant>" --review-status | Variant annotation in spatial context | all modes |
discover "<biological question>" | Concept resolution for question design | socrates |
enrich <gene1,gene2,...> | Gene-set enrichment for spatial clusters | all modes |
Usage rule: Always trigger BioMCP when the user asks about a specific gene, pathway, or disease context. For method discovery in Plato/Socrates modes, at least one BioMCP literature search is mandatory before recommending a pipeline.
Self-Evolution
This skill writes to ~/.spatial-transcriptomics-memory/ for persistent learning
across sessions:
- Findings log:
~/.spatial-transcriptomics-memory/findings.jsonl— append insights from each analysis session (methods that worked, edge cases, platform-specific quirks). - Method catalog:
reference/methods-catalog.md— curated list of spatial methods with platform compatibility, performance notes, and citation anchors. - Personality memory: Each mode file tracks its own session history in
~/.spatial-transcriptomics-memory/<mode>-sessions.jsonl(aristotle/plato/socrates) for context continuity.
Before starting any analysis, the skill should briefly scan the findings log for relevant past insights. After completing analysis, new findings are appended.
Code Templates
Reusable analysis templates are stored in templates/. These are Python scripts
or Jupyter notebook stubs that each mode can customize:
| Template | File | Description |
| QC & Preprocessing | templates/qc-template.py | Scanpy/Squidpy-based QC with configurable thresholds |
| Spatial Clustering | templates/clustering-template.py | Leiden/Louvain with spatial neighborhood graph |
| SV Gene Detection | templates/svg-template.py | SpatialDE, SPARK-X, Moran's I wrappers |
| Deconvolution | templates/deconv-template.py | Cell2location, RCTD, Tangram spot deconvolution |
| Niche Analysis | templates/niche-template.py | Spatial neighborhood composition analysis |
| Visualization | templates/viz-template.py | Publication-ready spatial plots (UMAP, spatial heatmaps) |
Each template includes placeholder comments (# TODO: CONFIGURE) where mode-specific
decisions are injected. Aristotle mode fills these with default values; Plato
and Socrates modes replace them with literature-backed or question-driven choices.
File Map
spatial-transcriptomics/
├── SKILL.md ← THIS FILE (router)
├── modes/
│ ├── aristotle.md ← Standard pipeline mode
│ ├── plato.md ← Literature-augmented mode
│ └── socrates.md ← Question-driven reverse design mode
├── reference/
│ └── methods-catalog.md ← Curated spatial methods reference
├── templates/
│ ├── qc-template.py
│ ├── clustering-template.py
│ ├── svg-template.py
│ ├── deconv-template.py
│ ├── niche-template.py
│ └── viz-template.py
└── memory/ ← Session-level scratch (not persisted)Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

