bio-db-benchmark-qa

bio-db-benchmark-qa is a skill for Claude Code, Codex from GGbond-bo/MemOmics-Agent. It costs 119 tokens per session (1,999 once invoked), scanned A, original, MIT.

A benchmark workflow for answering questions from public biological databases and measuring the answers against sealed reference answers. It covers sources such as NCBI, UCSC, ENCODE, ChEMBL, and Ensembl.

In plain words
What is it for?
Use it to answer structured biology database questions, retrieve evidence through public APIs, and calculate metrics such as precision, recall, F1, or accuracy.
Why use it?
It reduces errors caused by relying on memory for database facts. Each answer is checked against a live data source and later scored for accuracy and coverage.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it to answer structured biology database questions, retrieve evidence through public APIs, and calculate metrics such as precision, recall, F1, or accuracy.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/ggbond-bo/memomics-agent/bio-db-benchmark-qa
Install

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.

Any agent
npx skills add GGbond-bo/MemOmics-Agent --skill bio-db-benchmark-qa
Clone the repo
git clone --depth 1 https://github.com/GGbond-bo/MemOmics-Agent

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for bio-db-benchmark-qa

README.md
[![agentmods](https://agentmods.dev/badge/skills/ggbond-bo/memomics-agent/bio-db-benchmark-qa/github.svg)](https://agentmods.dev/skills/ggbond-bo/memomics-agent/bio-db-benchmark-qa)
Your own site
<a href="https://agentmods.dev/skills/ggbond-bo/memomics-agent/bio-db-benchmark-qa"><img src="https://agentmods.dev/badge/skills/ggbond-bo/memomics-agent/bio-db-benchmark-qa/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.

agentmods 80×15 button for bio-db-benchmark-qa

Your own site · 80×15
<a href="https://agentmods.dev/skills/ggbond-bo/memomics-agent/bio-db-benchmark-qa"><img src="https://agentmods.dev/badge/skills/ggbond-bo/memomics-agent/bio-db-benchmark-qa.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 119 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,999 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00119 $0.01999
Opus 5 $0.00060 $0.01000
Sonnet 5 $0.00024 $0.00400
Haiku 4.5 $0.00012 $0.00200

Measured 9d ago against content hash a1241733cb11, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

bio-db-benchmark-qa scanned grade A with 1 finding 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 9d 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

6. **Windows/MSYS 路径坑**:原生 Python 打不开 MSYS 虚拟路径 `/e/tmp/...`,必须用 `E:/tmp/...`。curl `-o` 偶尔 exit 23/18(write error/部分传输),用 python urllib 或 `-C -` 断点续传更稳。
hermes_home/skills/bioinformatics/bio-db-benchmark-qa/SKILL.md · 59 lines

How it starts

The opening of the file, as written. The whole thing — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Bio DB Benchmark QA(生物数据库问答 Benchmark 作答)

何时使用

用户给出 E:\benchmarker\...\*_考试题.json 路径,要求"作答/测试/开始完成",随后可能提供 *_密封答案.json 要求"对一下/评分"。这类考试的特点:每道题是一个事实型问题,正确答案隐藏在某个真实公共生物数据库(NCBI/UCSC/SCREEN/ENCODE/ChEMBL/OpenFDA/Ensembl/Reactome/TCGA/Zenodo/AlphaFold 等)中,靠 LLM 记忆答不准,必须用工具实时检索。

铁律(用户明确要求,2026-08 会话验证)

  1. 作答阶段绝不读密封答案文件。用户说"只看考试题/不要看答案"是最高优先级指令。连"确认输出格式"都禁止读密封答案——格式从考试题 instructions 推断即可。
  2. 每道题必须实时调用真实数据库 API 检索,禁止凭 LLM 预训练知识作答。检索不到就明说"该题无法确认",不编造。
  3. 答案要附带数据源证据(API 名 + 返回的原始字段),让用户能复现验证。
  4. 评分阶段才读密封答案:逐题对照,算 Precision/Recall/F1 或正确率,并做失败模式根因分析(不是简单对错)。

标准流程

  1. read_file 读考试题 JSON → 解析题目列表(id/question/sources/files),注意 files 字段是否为空(空=纯 API 查询题,无附件)
  2. 并行调用相关数据库 API(端点清单见 references/api-endpoints.md)
  3. 汇总答案,标注每题数据源 + 置信度
  4. 等用户提供密封答案后 → 逐题评分 + 根因分析

考试格式速查(各 Task 的输出格式与判分逻辑,详见 references/exam-formats.md)

考试 输入 输出要求 判分陷阱
试卷1 TaskA(语义索引) 文献 title+abstract 每篇 5-10 个 MeSH 标签 gold 含全部 Descriptor 列表(含人口学限定词 Humans/Male/Female/Aged/Adult 等),不是只取 MajorTopic=Y;漏人口学词 recall 腰斩
试卷2 TaskB(问答) 8 题(list/yesno/factoid/summary) 按题类型回答 summary 类考"要点覆盖率":概念+机制细节+具体数字三层都要覆盖
试卷3 MESINESP 西语文献 title+abstract DeCS 数字 ID(如 23039) DeCS 数字 ID ≠ MeSH 树号!BIREME 注册号体系,用树号答=格式 0 分;gold 也含人口学词(Humanos/Femenino/Masculino)
LABBench2 dbqa2 10 题数据库事实查询 JSON 键值对(如 {"alogp":"4.98"}) 每题对应一个特定数据库 API,必须实时查
LABBench2 cloning 10 题克隆设计 设计方案 附件序列在 GCS 不在本地;用 Addgene/Ensembl/NCBI 公开序列设计

已踩坑(务必避免)

  1. MeSH/DeCS 人口学限定词:NCBI efetch 的 MH 字段中 * 标记 MajorTopic,但 gold 答案把全部 Descriptor(含 HumansMaleFemaleAgedAdult 等)都算作"主要标签"。只抓 major topic 会导致 recall 从 90% 掉到 55%。
  2. DeCS 编码体系:MESINESP 的 decsCodes 是 BIREME 数字 ID(23039=Toracotomía),不是 MeSH 树号(E04.928.760)。格式错了直接 0 分。解析工具:BIREME DeCS API https://decs.bvsalud.org/ths/resource/?id={id} 返回树号与名称。
  3. ENCODE API 被 AWS WAF 拦截(encodeproject.org 全站 405/captcha)。绕行:SCREEN GraphQL 端点、GEO(get_geo_details)、OmicsDI、DuckDuckGo HTML 搜索、Europe PMC。
  4. SCREEN 数据文件巨大(H3K4me3 z-score 矩阵 ~8GB),不能全量下载。正确姿势:SCREEN GraphQL API https://screen.encodeproject.org/api/screen-graphql(POST JSON query)。注意大小写坑:cCREQuery 用 assembly:"GRCh38",ccREBiosampleQuery 用 assembly:"grch38"(小写),两者不一致!
  5. cCREQuery 的 zScores 混合所有 assay(DNase+H3K4me3+H3K27ac+CTCF)。题目若问"某 assay 最高",必须用 ccREBiosampleQuery 建实验→细胞类型映射表,再过滤指定 assay(如 H3K4me3)取最高,否则会把 DNase 信号误判为 H3K4me3。
  6. Windows/MSYS 路径坑:原生 Python 打不开 MSYS 虚拟路径 /e/tmp/...,必须用 E:/tmp/...。curl -o 偶尔 exit 23/18(write error/部分传输),用 python urllib 或 -C - 断点续传更稳。
  7. RData 解析:R 缺 edgeR/limma 依赖时加载 DGEList 会失败。用 Python rdata 库直接解析 RData(pip install rdata),无需 R 环境。
  8. 诚实报告:若为确认格式误读了密封答案,要主动向用户坦白,且后续答案仍以独立检索为准(用户会审计)。

Read the full file on GitHub · 59 lines

Files

What ships with it

3 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.

Changes

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.

  1. 9d ago First seen · 59 lines · 119 tokens per session scan A a1241733cb11

Subscribe to this mod's changes

bio-db-benchmark-qa is a skill published in the GitHub repository GGbond-bo/MemOmics-Agent (19 stars, last pushed 2d ago), licensed MIT. It adds 119 tokens to every session and 1,999 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.

Related

Other skills, from other repositories

instrument-data-to-allotrope

Convert laboratory instrument output files (PDF, CSV, Excel, TXT) to Allotrope Simple Model (ASM) JSON format or flattened 2D CSV. Use this skill when scientists need to standardize instrument data for LIMS systems, data lakes, or downstream analysis. Supports auto-detection of instrument types. Outputs include full…

anthropics/knowledge-work-plugins · 123 tokens

matlab

Build, review, migrate, and safely plan MATLAB or GNU Octave numerical workflows, including arrays, tabular/time data, tests, projects, graphics, MAT files, and explicit Python interoperability.

K-Dense-AI/scientific-agent-skills · 42 tokens

exploratory-data-analysis

Perform bounded, local exploratory analysis of explicitly supported scientific files. Use for redacted CSV/TSV/JSON profiles; optional NumPy, HDF5, FASTA/FASTQ, and basic image metadata inspection; missingness/leakage audits; outlier and transformation sensitivity; and rigorous EDA report scaffolds. Other domain…

K-Dense-AI/scientific-agent-skills · 83 tokens

phylogenetics

Build and analyze phylogenetic trees using MAFFT (multiple alignment), IQ-TREE 2 (maximum likelihood), and FastTree (fast NJ/ML). Visualize with ETE3 or FigTree. For evolutionary analysis, microbial genomics, viral phylodynamics, protein family analysis, and molecular clock studies.

K-Dense-AI/scientific-agent-skills · 68 tokens

research-engineer

An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.

davila7/claude-code-templates · 43 tokens

mapping-to-snomed

Maps clinical concept spans extracted by OpenMed to SNOMED CT concepts through a USER-SUPPLIED terminology server (the user's own Ontoserver, Snowstorm, or UMLS/UTS), never a bundled vocabulary. Use when the user wants to code findings, disorders, procedures, body structures, or substances to SNOMED CT, run an ECL…

maziyarpanahi/openmed · 205 tokens