Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add TianGzlab/OmicsClaw --skill sc-perturb-prepgit clone --depth 1 https://github.com/TianGzlab/OmicsClawWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/tiangzlab/omicsclaw/sc-perturb-prep)<a href="https://agentmods.dev/skills/tiangzlab/omicsclaw/sc-perturb-prep"><img src="https://agentmods.dev/badge/skills/tiangzlab/omicsclaw/sc-perturb-prep/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/tiangzlab/omicsclaw/sc-perturb-prep"><img src="https://agentmods.dev/badge/skills/tiangzlab/omicsclaw/sc-perturb-prep.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Rogue Agent · line 3 Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.Fix: Prevent the skill from modifying its own code, SKILL.md, or configuration files. Treat skill files as read-only at runtime.
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00085 | $0.01176 |
| Opus 5.5 | $0.00034 | $0.00470 |
| Sonnet 5.5 | $0.00017 | $0.00235 |
| Haiku 4.5 | $0.00009 | $0.00118 |
Grade A, and why
sc-perturb-prep scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 20d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.
sc-perturb-prep
Attach an upstream barcode-to-guide table to expression data. This is the
mapping_tsv method, not a FASTQ guide caller. It needs no pertpy installation.
Key CLI
python skills/singlecell/scrna/sc-perturb-prep/sc_perturb_prep.py --demo --output /tmp/sc_perturb_prep_demo
python skills/singlecell/scrna/sc-perturb-prep/sc_perturb_prep.py --input expression.h5ad --mapping-file mapping.tsv --output results/prep
Mapping columns are inferred from common names, or selected with
--barcode-column, --sgrna-column, and --target-column. Without target
values, --delimiter _ --gene-position 0 extracts the target from a guide ID.
Pass --keep-multi-guide only when retaining multi-guide cells is intended.
Workflow
Upstream: expression counts and guide calls from the same cells. Standardize
mapping columns, collapse guides per barcode, match assigned cells, retain
gene-expression features and canonicalize the matrix. Downstream: use
sc-perturb for observed perturbation signatures, or sc-preprocessing for
clustering. The API returns objects; the example step writes them explicitly.
Matrix Contract
Raw counts are preferred; canonicalization can recover counts from a layer or
raw snapshot. Inspect omicsclaw_matrix_contract and run_info for the actual
expression source. A normalized matrix is not evidence of raw counts.
Inputs & Outputs
Input: .h5ad, 10x H5 or a matrix directory, plus mapping TSV/CSV for real runs.
The CLI writes processed.h5ad, report.md, result.json,
reproducibility/commands.sh, tables/perturbation_assignments.csv,
tables/assignment_status_counts.csv, tables/perturbation_counts.csv,
tables/feature_type_summary.csv and figures/perturbation_counts.png.
tables/dropped_multi_guide_cells.csv is written only when rows were dropped.
Plot data are also written under figure_data/.
Gotchas
tables/perturbation_assignments.csvuses statusassigned,controlor retainedmulti_guide; dropped rows do not appear in the output AnnData.- Control tokens match whole words separated by punctuation:
NT_sg1is a control, whileWNT3_sg1,NTRK1_sg1andNT5E_sg1are not. Reviewtables/perturbation_assignments.csvafter choosing custom patterns. result.json→summary.n_cells_multi_guide_droppedcounts dropped multi-guide cells. No matching barcodes raisesValueErrorin the API.tables/feature_type_summary.csvdescribes the input feature types. Filtering needsvar['feature_types']; absent labels cannot identify guide or antibody features.
What ships with it
9 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- _api.py 4.9 KB runs code
- examples/example_step.py 1.4 KB runs code
- references/methodology.md 719 B
- references/output_contract.md 537 B
- references/parameters.md 476 B
- references/r_visualization.md 288 B
- sc_perturb_prep.py 18 KB runs code
- tests/test_prep_api.py 3.0 KB runs code
- tests/test_sc_perturb_prep_methods.py 1.8 KB runs code
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 20d ago Changed · -14 lines 1f2975c48468
- 1mo ago First seen · 124 lines · 85 tokens per session scan A a7a44880d343
sc-perturb-prep is a skill published in the GitHub repository TianGzlab/OmicsClaw (161 stars, last pushed 20d ago), licensed Apache-2.0. It adds 85 tokens to every session and 1,176 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
scanpy
Standard single-cell RNA-seq analysis pipeline. Use for QC, normalization, dimensionality reduction (PCA/UMAP/t-SNE), clustering, differential expression, and visualization. Best for exploratory scRNA-seq analysis with established workflows. For deep learning models use scvi-tools; for data format questions use…
Virtual Embryo — atlas data + knowledge graph
Query the Virtual Embryo knowledge graph (mouse/human developmental biology: genes, anatomy, Theiler/Carnegie stages, gene expression, diseases, papers) and its 3D atlas catalog (anatomical OPT/light-sheet volumes + 3D spatial- transcriptomics datasets), and visualise those datasets in 3D with the volume3d / spatial3d…
Single-Cell Analysis Skills Index
Core skills for single-cell RNA-seq analysis: quality control, cell type annotation, and trajectory inference. These are high-priority actionable workflows — load them first for common single-cell tasks.
cellxgene-census-query
Query CZ CELLxGENE Census (61M+ cells). Filter by cell type/tissue/disease, retrieve expression data, and integrate with scanpy/PyTorch for population-scale single-cell analysis. Use this skill when: (1) Querying single-cell expression data by cell type, tissue, or disease, (2) Exploring available single-cell datasets…
single-cell-scrna-seq-analysis-scanpy
Complete single-cell RNA-seq analysis workflow built on Scanpy and AnnData. Use this skill when: (1) Loading diverse single-cell data formats (10X, h5ad, CSV), (2) Performing quality control and filtering, (3) Normalization, dimensionality reduction, and clustering, (4) Marker gene identification and cell type…
single-cell-multi-omics-analysis-scvi
Probabilistic deep learning framework for single-cell multi-omics data analysis. Use this skill when: (1) Analyzing single-cell RNA-seq data with batch correction, (2) Integrating multi-modal data (CITE-seq, ATAC-seq, multi-omics), (3) Performing cell type annotation with scANVI, (4) Spatial transcriptomics…