Spatial Annotate

What it does

When your agent uses it

• Tasks that involve Bioinformatics

Example prompts

• “/spatial-annotate”

Requirements

• Python 3

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

1. 1Load AnnData (--input) or chain through spatial-preprocess --demo.
2. 2parser.error validates per-method numeric flags (lines :730-746); :748 raises if --reference is missing for tangram / scanvi; :750 for…
3. 3Dispatch to method
4. 4Write obs["cell_type"] (Categorical); per-method extras (probabilities, marker overlap).
5. 5Save processed.h5ad, tables, figures, report.md, result.json.

What it can do on your machine

Read from SKILL.md and the folder at commit 6fbd79f. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Ships script files (Python and R), which the agent can run.

  Shell commands in SKILL.md call:

  • python

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Spatial Annotate loads about 1.8k tokens when it runs, and up to ~6.2k if it reads all its reference files. Until then it costs about 81 tokens; SKILL.md has 472 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~81

When it runs · the whole SKILL.md, loaded when a task matches

~1.8k

With references · SKILL.md plus every file in references/, read only if the agent opens them

~6.2k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from TianGzlab/OmicsClaw at commit 6fbd79f, republished under its MIT licence (© TianGzlab). 472 words, ~1,751 tokens.

name

spatial-annotate

description

Load when assigning per-spot cell-type labels on a spatial AnnData via marker-gene scoring or scRNA-reference mapping (Tangram / scANVI / CellAssign). Skip when computing spot-level cell-type proportions for multi-cell-per-spot platforms (use spatial-deconv); tissue-domain detection (use spatial-domains).

version

0.5.0

author

OmicsClaw

license

MIT

emoji

🏷️

tags

spatial, annotation, cell-type, marker-based, tangram, scanvi, cellassign

requires

anndata, matplotlib, numpy, pandas, scanpy, scipy, scvi-tools, seaborn, tangram-sc, torch

<!-- AUTO-GENERATED from skill.yaml (interface) — do not edit by hand. Regenerate: python scripts/generate_skill_md.py <skill_dir> -->

# Demo (chained from spatial-preprocess --demo, marker_based)
python omicsclaw.py run spatial-annotate --demo --output /tmp/spatial_annot_demo

# Marker-based on a Visium with built-in human markers
python omicsclaw.py run spatial-annotate \
  --input clustered.h5ad --output results/ \
  --method marker_based --species human --marker-n-genes 50

# Tangram reference mapping
python omicsclaw.py run spatial-annotate \
  --input clustered.h5ad --output results/ \
  --method tangram --reference scrna_atlas.h5ad \
  --tangram-num-epochs 1000 --tangram-train-genes 1000

# scANVI semi-supervised
python omicsclaw.py run spatial-annotate \
  --input clustered.h5ad --output results/ \
  --method scanvi --reference scrna_atlas.h5ad \
  --scanvi-n-latent 30 --scanvi-max-epochs 400 --batch-key sample

# CellAssign with marker matrix
python omicsclaw.py run spatial-annotate \
  --input clustered.h5ad --output results/ \
  --method cellassign --cellassign-max-epochs 200

© TianGzlab, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 8 other files (references) in skills/spatial/spatial-annotate of TianGzlab/OmicsClaw.

• SKILL.md
• r_visualization/README.md
• r_visualization/annotation_publication_template.R
• references/methodology.md
• references/output_contract.md
• references/parameters.md
• skill.yaml
• spatial_annotate.py
• tests/test_spatial_annotate.py

Open the folder on GitHubat commit 6fbd79f