← Back to Skills Marketplace
307
Downloads
0
Stars
0
Active Installs
2
Versions
Install in OpenClaw
/install ncbi-search
Description
Search NCBI databases using E-Utilities API (official, free). Supports multiple databases: PubMed (literature), Gene, Protein, Nucleotide, dbSNP, ClinVar, Ta...
Usage Guidance
This skill appears to do what it says: local Python scripts call NCBI E-Utilities and optionally use your NCBI API key for higher rate limits. Before installing: 1) Verify you have Python and install the 'requests' package (README shows pip install requests) — SKILL.md's "no external dependencies" claim is inaccurate. 2) Confirm you trust the skill source (source/homepage unknown). 3) If you plan heavy usage, create and set an NCBI API key to avoid throttling. 4) Review the included scripts yourself (they are short and readable) or run them in a sandboxed environment first. Finally, note the skill will make outbound network calls to NCBI (expected), but it does not request other credentials or access system-wide secrets.
Capability Analysis
Type: OpenClaw Skill
Name: ncbi-search
Version: 1.0.1
The ncbi-search skill bundle is a well-implemented tool for querying official NCBI databases (PubMed, Gene, dbSNP, etc.) via the E-Utilities API. The Python scripts (ncbi_search.py, pubmed_search.py) use the standard requests library to interact with official NCBI endpoints, incorporating appropriate rate limiting and retry logic. There is no evidence of data exfiltration, malicious execution, or prompt injection; the instructions in SKILL.md are strictly aligned with the stated purpose of biomedical data retrieval.
Capability Assessment
Purpose & Capability
Name/description match the included code: scripts call NCBI E-Utilities (esearch, esummary, efetch) and implement intent detection for PubMed, Gene, Protein, dbSNP, ClinVar, taxonomy, etc. No unrelated services or credentials are requested.
Instruction Scope
SKILL.md instructs running the provided Python scripts and to set NCBI_API_KEY (optional). Instructions reference only user-specified files (PMID lists) and NCBI endpoints. Minor issues: SKILL.md/README claim "no external dependencies" while the code requires the third-party 'requests' library and the README shows pip install requests; there's also a small path typo in one example ("~/.agents/skill/" vs "~/.agents/skills/"). These are implementation/documentation inconsistencies but not malicious scope creep.
Install Mechanism
No install spec in the registry (instruction-only install). The project includes Python scripts and expects the 'requests' package; README suggests pip install requests. No downloads from arbitrary URLs or archive extraction are present. The lack of an explicit install step means users must manually ensure Python and dependencies are available.
Credentials
The only credential discussed is the optional NCBI_API_KEY (used to raise rate limits) and it is accessed via command-line arg or NCBI_API_KEY environment variable. No other secrets, unrelated environment variables, or config paths are requested.
Persistence & Privilege
Registry flags are default (always:false, user-invocable:true). The skill does not request persistent elevated privileges or modify other skills. It will perform network requests to NCBI when invoked; no autonomous always-on behavior is requested.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install ncbi-search - After installation, invoke the skill by name or use
/ncbi-search - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
- Initial release of the ncbi-search skill.
- Search multiple NCBI databases (PubMed, Gene, Protein, Nucleotide, dbSNP, ClinVar, Taxonomy, etc.) using the official E-Utilities API.
- Automatically detects the query's intent and selects the most appropriate database.
- Returns formatted results with key information for biomedical literature, gene/protein information, variants, and more.
- Supports both automatic search mode and manual database selection.
- Optional: Configure NCBI API Key for higher query rates.
v1.0.0
- Initial release of the ncbi-search skill.
- Search across multiple NCBI databases (PubMed, Gene, Protein, Nucleotide, dbSNP, ClinVar, Taxonomy, and more) via the official E-Utilities API.
- Supports smart query intent recognition to automatically select the appropriate database.
- Provides formatted search results with key information for literature, genes, proteins, variants, and other biomedical data.
- Allows database specification, filters (years, article type), organism restriction, and API key configuration for higher rate limits.
- Includes usage instructions, sample outputs, and references.
Metadata
Frequently Asked Questions
What is ncbi-search?
Search NCBI databases using E-Utilities API (official, free). Supports multiple databases: PubMed (literature), Gene, Protein, Nucleotide, dbSNP, ClinVar, Ta... It is an AI Agent Skill for Claude Code / OpenClaw, with 307 downloads so far.
How do I install ncbi-search?
Run "/install ncbi-search" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is ncbi-search free?
Yes, ncbi-search is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does ncbi-search support?
ncbi-search is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created ncbi-search?
It is built and maintained by Pengside (@side-peng); the current version is v1.0.1.
More Skills