Skip to main content
Glama
kmaneesh

BioPython MCP Server

by kmaneesh

pubmed_review

Generate a formatted literature review from PubMed search results, including complete abstracts, and write as a Markdown file directly to your Obsidian vault or specified directory.

Instructions

Create a formatted literature review from PubMed search results and write to MD file.

This function searches PubMed, fetches article metadata, formats it as markdown with complete abstracts, and writes the content directly to a file. The LLM determines both the storage location and filename.

Args: query: PubMed search query (supports full Entrez syntax including year filters) Example: "BRCA1 AND breast cancer AND 2020:2024[PDAT]" path: Relative directory path within vault (e.g., "research/cancer" or "genetics/reviews") The LLM decides the directory structure. filename: Name of the markdown file (e.g., "brca1_review_2024.md" or "alport_syndrome.md") The LLM decides the filename. Should include .md extension. obsidian_vault: Path to Obsidian vault root (optional, defaults to OBSIDIAN_VAULT_PATH env variable) Example: "/Users/user/Documents/MyVault" max_results: Maximum number of articles to include (default: 25, max: 1000) sort: Sort order - "pub_date", "relevance", etc. (default: "pub_date")

Returns: Dictionary with review results and metadata: - status: "success" or "error" - filepath: Full path where file was written - articles_found: Total number of articles found - articles_written: Number of articles successfully processed - articles_with_pmc: Count of articles with PMC IDs - articles_with_doi: Count of articles with DOIs - query: Original search query - file_size_kb: File size in kilobytes - year_range: {"min": int, "max": int} - top_journals: List of top 5 journals by article count - execution_time_seconds: Time taken to generate review

Examples: >>> # Using environment variable for vault path >>> result = pubmed_review( ... query="COL4A3[Gene] AND Alport syndrome", ... path="genetics/reviews", ... filename="alport_syndrome_2024.md" ... )

>>> # Overriding vault path
>>> result = pubmed_review(
...     query="BRCA1 AND breast cancer AND 2020:2024[PDAT]",
...     path="oncology/brca1",
...     filename="literature_review_jan2024.md",
...     obsidian_vault="/Users/user/Vault",
...     max_results=50
... )

Notes: - Vault path comes from OBSIDIAN_VAULT_PATH environment variable or obsidian_vault parameter - LLM controls both directory structure (path) and filename separately - Writes markdown file directly to disk with complete abstracts - Fetches articles in batches of 20 (NCBI limit) - Respects NCBI rate limits (3/sec or 10/sec with API key) - For very large reviews (>500 articles), consider splitting into multiple calls - Includes Obsidian-compatible YAML frontmatter

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
queryYes
pathYes
filenameYes
obsidian_vaultNo
max_resultsNo
sortNopub_date

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault

No arguments

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries full burden. It discloses key behaviors: writes to disk, fetches in batches of 20, respects NCBI rate limits, includes YAML frontmatter, and uses environment variables for vault path. However, it does not state whether existing files are overwritten or what happens on write failure, leaving a minor gap.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is well-structured with clear sections (Args, Returns, Examples, Notes) and uses markdown formatting. It is somewhat long (approx. 30 lines) but each part adds value. No redundant sentences; every section earns its place. Minor improvement could be trimming common knowledge (e.g., 'file should include .md extension' is implied).

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's complexity (6 parameters, output dictionary, sibling differentiation), the description is complete. It covers purpose, all parameters with examples, return value schema, behavioral notes (rate limits, batching), and environmental context. The presence of a detailed Returns section means no need to explain return values further. No gaps identified.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Since input schema has 0% description coverage, the description entirely defines parameter semantics. It provides detailed explanations for all 6 parameters, including examples (e.g., query syntax), defaults, constraints (max_results max 1000), and valid options (sort values). This fully compensates for the schema's lack of descriptions.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states it creates a formatted literature review from PubMed and writes to an MD file. The verb 'Create', resource 'literature review from PubMed', and output 'write to MD file' are specific. This distinguishes it from sibling tools like pubmed_search which only return results, or entrez_fetch which returns raw data.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides context about LLM-controlled location and file writing, and adds a note about splitting large reviews. However, it does not explicitly state when to use this tool vs alternatives (e.g., pubmed_search for simple queries, entrez_fetch for raw data), nor does it include 'when not to use' guidance. Usage is implied but not explicitly bounded.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/kmaneesh/biopython-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server