跳到主要内容
知知仓学习社ZHICANG

cellagent-annotation

cellagent-annotation，来自 FreedomIntelligence/OpenClaw-Medical-Skills 的 agent 技能。

执行命令联网写文件严重 0 · 高危 8FreedomIntelligence/OpenClaw-Medical-Skills

它会碰到什么

扫了多少176 个文本文件，2593 KB
它会碰到什么执行命令联网写文件
命中总数19 处
命中统计严重 0 · 高 8 · 中 5 · 低 6
逐条看命中（8 条严重或高危）
  • 高 repo/CellTypeAgent/get_prediction.py:122exec-spawn
    pred = eval(data_with_prediction.loc[i, 'cell_type_pred']) if isinstance(
  • 高 repo/CellTypeAgent/get_selection.py:161exec-spawn
    cell_type_pred_CLname = eval(sample['cell_type_pred_CLname']) if isinstance(sample['cell_type_pred_CLname'], str) else sample['cell_type_pred_CLname']
  • 高 repo/CellTypeAgent/get_selection.py:475exec-spawn
    pred_ls = pred[pred_name].apply(lambda x: eval(x) if isinstance(x, str) else x).tolist()
  • 高 repo/CellTypeAgent/get_selection.py:525exec-spawn
    lambda x: eval(x) if isinstance(x, str) and '[[' in x else ([ [x] ] if isinstance(x, str) else x)
  • 高 repo/CellTypeAgent/get_selection.py:551exec-spawn
    lambda x: eval(x) if isinstance(x, str) and '[[' in x else ([ [x] ] if isinstance(x, str) else x)
  • 高 repo/CellTypeAgent/get_selection.py:554exec-spawn
    lambda x: eval(x) if isinstance(x, str) and '[[' in x else ([ [x] ] if isinstance(x, str) else x)
  • 高 repo/CellTypeAgent/get_selection.py:572exec-spawn
    sep_avg_score += eval(data[f'final_score_{j}'][i]) if isinstance(data[f'final_score_{j}'][i], str) else data[f'final_score_{j}'][i]
  • 高 repo/CellTypeAgent/get_selection.py:581exec-spawn
    data['best_candidate_index'] = data['agreement_score'].apply(lambda x: np.argmax(eval(x)))

这一栏是扫描器报的事实，不是结论。命中多不等于有毒（安全工具、规则库、示例脚本本来就会包含危险写法），命中少也不等于干净。它和你手上的凭据、文件、网络有什么关系，需要你自己看。

技能内容

<!--

COPYRIGHT NOTICE

This file is part of the "Universal Biomedical Skills" project.

Copyright (c) 2026 MD BABU MIA, PhD <md.babu.mia@mssm.edu>

All Rights Reserved.

#

This code is proprietary and confidential.

Unauthorized copying of this file, via any medium is strictly prohibited.

#

Provenance: Authenticated by MD BABU MIA

-->

name: cellagent-annotation

description: Cell tagger

keywords:

  • single-cell
  • markers
  • annotation
  • confidence
  • tissue

measurable_outcome: Label every provided cluster with a cell type + confidence + marker evidence (or "ambiguous") within 15 minutes per dataset.

license: MIT

metadata:

author: CellAgent Team

version: "1.0.0"

compatibility:

  • system: Python 3.9+

allowed-tools:

  • run_shell_command
  • read_file

CellAgent Annotation

Use CellTypeAgent to interpret marker genes, annotate scRNA-seq clusters, and coordinate multi-agent workflows for downstream analysis.

When to Use

  • Automated annotation of scRNA-seq datasets without manual curation.
  • Multi-step workflows (QC → clustering → annotation → DE analysis).
  • Integrating multiple batches requiring consistent labeling.

Core Capabilities

  1. Planning: Multi-agent planner decomposes analysis goals into steps.
  2. Tool execution: Generates Scanpy/Seurat code and runs it autonomously.
  3. Self-correction: Detects execution errors and retries with fixes.

Workflow

  1. Gather marker lists per cluster, plus species/tissue context and optional atlas references.
  2. Run CellTypeAgent (pip install -r requirements.txt then python repo/main.py --data data.h5ad --goal annotate).
  3. Review outputs for supporting markers; downgrade ambiguous clusters when signals conflict.
  4. Produce final table (cluster, label, confidence, supporting markers, notes) and cite references when used.

Example Usage

python3 Skills/Genomics/Single_Cell/CellAgent/repo/main.py --data "./data.h5ad" --goal "annotate"

Guardrails

  • Avoid over-specific lineages if markers overlap; default to broader types.
  • Flag clusters showing multiple signatures for manual review.
  • Respect species/tissue differences when interpreting markers.

References

  • README + upstream paper (Mao et al., 2025 / arXiv 2407.09811).

<!-- AUTHOR_SIGNATURE: 9a7f3c2e-MD-BABU-MIA-2026-MSSM-SECURE -->

想直接用这个技能？

本站把开放许可（MIT / Apache 等）的技能按仓库打包整理到网盘，点一下转存到你自己的网盘，不用一个个从 GitHub 拉。许可未声明的技能只给原始仓库链接，不打包。

去原始仓库

它属于哪个仓库

仓库FreedomIntelligence/OpenClaw-Medical-Skills
星标★ 3,010
本站分层T1
该仓技能数897
原文件路径skills/cellagent-annotation/SKILL.md

同一个仓库里的其他技能

看这个仓库的全部 897 个技能