paper-fetch-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
Known papers only: DOI, URL or title; resolve ambiguity before fetching. Honor user files/sources, then valid session handles, then qualified local text. Use batch_fetch for multiple papers. Shared fetch presets: temporary_read (default) returns Markdown handles without paper/asset/cache writes; cacheable_read reuses/writes request-matching cache in download_dir or configured scope; archive saves Markdown in markdown_output_dir or download_dir, plus attachments/index for target=supplementary. Required paths must be supplied. Explicit options, including nested strategy fields and false values, override presets. Defaults: full_text, all references, metadata fallback, no body assets. body/all assets enable downloads and artifact_mode=markdown-assets unless explicitly overridden; they may add bounded ImageContent. Runtime preparation is independent of paper writes and on demand. Business payloads: schema_version=4 only in structuredContent; request article explicitly. Saved files appear in output_artifacts; body content stays in content_ref/read_content; navigate_content locates headings/literal matches. Judge acquisition by acceptance, text by pagination/truncation, reading completion by consumed portions. Provider/platform facts: resource://paper-fetch/provider-catalog. Respect access controls and manual-auth boundaries.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| resolve_paperA | Resolve a DOI, URL, or title query into a normalized paper candidate. |
| fetch_paperB | Fetch one known paper, or materials only with target=supplementary. Returns metadata, acceptance and content_ref for read_content, also after saving. Shared presets control cache/files; temporary_read writes neither. |
| read_contentA | Read retained Markdown/article/diagnostics locally without writes; expired handles never refetch. Offsets are Unicode code points; continue with next_offset. eof ends this view, not necessarily the untruncated original. |
| navigate_contentC | Navigate retained Markdown headings or case-insensitive literal matches. Ranges use original Unicode code points for read_content; never refetches. |
| list_cachedB | List DOI-proven downloads in the selected cache index; no network or loose-file scan. |
| get_cachedB | Read DOI-proven cache entries in one download_dir without network/writes. compact gives preferred entries and request-sensitive acceptance/assets; preferred_only omits other entry arrays. |
| batch_resolveB | Resolve multiple DOI, URL or title queries with optional concurrency. |
| batch_fetchB | Batch fetch papers or supplementary materials using fetch_paper presets and cache/file options. Returns input-ordered metadata, acceptance and content_ref for read_content; aliases share acquisition/handles. batch_results opts into an atomic JSONL manifest. |
| batch_checkA | Probe availability from metadata/landing signals; no full-body fetching or download writes. Input-ordered results contain probe_state, evidence and errors. Use batch_fetch for acquisition. |
| browser_preflightB | Live-check provider HTML with the managed browser runtime; opens publisher pages and updates profiles shared with auth/fetch. Exports JSON diagnostics; no PDF fallback or automatic authentication. |
| provider_statusA | Inspect static provider configuration and local dependencies without network/browser access. Use browser_preflight for live checks. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| provider_catalog | Runtime-derived provider, source, browser/preflight, and asset-default catalog. |
TDQS
Scored across 11 tools
Tools are clearly separated by action and scope: resolve vs batch_resolve, fetch vs batch_fetch, read_content vs navigate_content, and provider_status vs browser_preflight. The cache tools list_cached and get_cached overlap slightly but descriptions distinguish index listing from entry reading.
Most tools follow a readable action_object convention such as resolve_paper, fetch_paper, read_content, and navigate_content, with a consistent batch_* grouping. Minor deviations like provider_status and browser_preflight use noun-noun forms, but overall the convention is predictable.
At 11 tools, the surface is well-scoped for paper resolution, fetching, reading, caching, batch operations, and diagnostics. Each tool covers a distinct part of the workflow without excessive redundancy.
The core lifecycle is covered: resolve, check availability, fetch, read/navigate content, list/read cache, and inspect providers. Minor gaps exist around explicit authentication and cache management or cleanup, but agents can work around these for typical workflows.