Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
pubmed-edirect 0.4.4 - Improved security advisory and installation instructions in documentation - Enhanced clarity in descriptions and instructions for manual skill setup - Added/updated security best practices and removed duplicated content - Standardized environment variable descriptions and removed non-English sections
What This Skill Does
Enables searching and retrieving biomedical literature from PubMed and other NCBI databases using the EDirect command-line utilities. Requires manual installation of the EDirect toolset on the user's system. Supports cross-database linking, batch processing, and structured XML extraction via Unix pipelines.
EDirect gives direct programmatic access to all NCBI databases through composable Unix pipes, avoiding the limitations of the web interface for batch or automated queries.
When to use it
- Fetch abstracts for a list of PMIDs in bulk
- Export PubMed search results to CSV with metadata
- Analyze publication trends over a date range
- Search gene or protein sequence databases from the terminal
- Cross-link records between PubMed and NCBI Gene database
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: find the 20 most recent abstracts on CRISPR gene editing and save them to a file
- 1Run esearch to query PubMed with the term CRISPR[TIAB] and set retmax to 20
- 2Pipe results through efetch with format set to abstract
- 3Use xtract or redirect output to parse and save abstracts to a text file
- 4Confirm the output file contains the expected number of records
A text file containing 20 recent PubMed abstracts on CRISPR gene editing
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.