Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Single-Cell QC · xuzhougeng | DecimalAI

scienceby xuzhougeng

Single-Cell QC

xuzhougeng/singlecell-qc

Source xuzhougeng/wisp-science ↗License AGPL-3.0

Use when designing, reviewing, or implementing single-cell RNA-seq QC in Python or R with a human-in-the-loop, data-driven approach. Trigger for scRNA QC metrics, per-sample diagnosis, threshold discussion, mitochondrial/ambient/doublet assessment, MAD vs fixed cutoffs, or refactoring automated merge-first QC. The analyst confirms key decisions at each step—agents must inspect data, propose options, and wait for approval before filtering, doublet removal, or merging. Not a turnkey pipeline skill

SkillScore32Maintained

Automatic

Data Analyst pack

Last verified 9/24/2026

What it changes

Passed 4 of 13 test cases with the skill, and 3 of 13 without it.

Without skill23%

With skill32%

Checks it passed that the model alone missed

Asks the user to confirm the chosen decontamination method before running.
Exports calculated metadata to a tabular file for diagnostic figure plotting.
Acknowledges legacy atlas reproduction as an allowed exception to the default per-sample human gate rule.

and 1 more in the full report

3 fixed4 passed either way14 still failing1 broke

Pass-rate lift: +9 pp
Tokens: +41%
Agent turns: 0%
Response time: -32%

Model gemini-3.6-flashMeasured 9/24/2026v17 of 22 cases passed with skill · 13 comparable

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/xuzhougeng-singlecell-qc/SKILL.md https://app.decimal.ai/s/xuzhougeng-singlecell-qc/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add xuzhougeng-singlecell-qc, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/xuzhougeng-singlecell-qc/ and pins it in .decimal/skills.lock

decimalai skills add xuzhougeng-singlecell-qc

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/xuzhougeng-singlecell-qc/SKILL.md https://app.decimal.ai/s/xuzhougeng-singlecell-qc/SKILL.md

Raw file: /s/xuzhougeng-singlecell-qc/SKILL.md ↗ · machine-readable summary at /s/xuzhougeng-singlecell-qc.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 3 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 3

Prompt

We are writing `01-calculate_metrics.R` using Seurat for `data/liver.rds`. What core metrics should be added to `[email protected]` before generating diagnostic plots?

What was checked, without and with the skill
Checked	Without skill	With skill
Assigns core metrics using consistent schema keys `n_genes`, `n_UMIs`, and `mito_frac` in Seurat metadata.	Missed	Passed
Exports calculated metadata to a tabular file for diagnostic figure plotting.	Missed	Passed

case-18 · gemini-3.6-flash · 9/24/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md158 lines11 filesincludes scripts

Single-Cell QC

Overview

Use this skill for data-driven, human-centered single-cell QC. The analyst inspects distributions and confirms decisions; code computes metrics and executes only what was agreed.

inspect data → compute metrics → human reviews → confirm thresholds → small action → re-inspect

Not a one-click pipeline. Do not chain calculate → filter → doublet → merge unless the user explicitly requests full execution after reviewing pilot results.

Follow analysis-workflow for module and script layout. Match the user's language.

Read first: references/human-in-the-loop.md

When To Use

"帮我看看这个样本 QC"
"算一下 QC 指标，阈值我来定"
"逐样本诊断，先别过滤"
"这个 merge-first QC 太粗，怎么改成人工确认"
"参考 GZL metrics 脚本，但要分步做"

Do not use for integration/Harmony, annotation, or spatial QC unless only expression-matrix QC is needed.

Operating Rules (Human-First)

Inspect before acting — matrix type, species, sample metadata, existing checkpoints.
Pilot samples first — default 1–3 samples; expand only after user OK.
Metrics before filters — run 01-calculate_metrics; stop and report.
Propose thresholds, never silently apply — show expected cell loss per sample.
Ask at gates — which metrics next? which thresholds? proceed to filter? merge?
No silent heavy steps — no full-cohort filter, Scrublet, decontX, or merge without explicit approval.
Reversible checkpoints — pre-filter metadata/counts stay intact; filtering writes new files.
Scripts = one stage — owner-editable; thresholds visible at top of filter scripts.

Full gate definitions: references/human-in-the-loop.md

First Pass (Always)

Track record

Benchmark

4 / 13

test cases passed, vs 3 / 13 without

Adoption

Users: 0
Times used: 0

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v5 — 8/17/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

xuzhougeng / wisp-science

skills/singlecell-qc/SKILL.md

View on GitHub ↗

License AGPL-3.0Imported 8/17/2026Version v1Bundled 10 files · scripts

More skills from the same source and nearby in the registry, with their headline benchmark lift.

Ready to supercharge your AI agents?

See it in action

What's inside

Single-Cell QC

Overview

When To Use

Operating Rules (Human-First)

First Pass (Always)

Track record

Benchmark

Adoption

SkillSafety

Source

Related skills