paper-fetch

name	paper-fetch
description	Use when the user wants to download a paper PDF from a DOI, title, or URL via legal open-access sources. Tries Unpaywall, arXiv, bioRxiv/medRxiv, PubMed Central, and Semantic Scholar in order. Never uses Sci-Hub or paywall bypass.
homepage	https://github.com/Agents365-ai/paper-fetch
metadata	{"openclaw":{"requires":{"bins":"[Truncated]"},"emoji":"📄"},"pimo":{"category":"research","tags":["paper","pdf","doi","open-access","download"]}}

paper-fetch

Fetch the legal open-access PDF for a paper given a DOI (or title). Tries multiple OA sources in priority order and stops at the first hit.

Agent-native. Structured JSON envelope on stdout, NDJSON progress on stderr, stable exit codes, machine-readable schema, TTY-aware format default, idempotent retries.

Resolution order

1. Unpaywall — https://api.unpaywall.org/v2/{doi}?email=$UNPAYWALL_EMAIL, read best_oa_location.url_for_pdf (skipped if UNPAYWALL_EMAIL not set)
2. Semantic Scholar — https://api.semanticscholar.org/graph/v1/paper/DOI:{doi}?fields=openAccessPdf,externalIds
3. arXiv — if externalIds.ArXiv present, https://arxiv.org/pdf/{arxiv_id}.pdf
4. PubMed Central OA — if PMCID present, https://www.ncbi.nlm.nih.gov/pmc/articles/{pmcid}/pdf/
5. bioRxiv / medRxiv — if DOI prefix is 10.1101, query https://api.biorxiv.org/details/{server}/{doi} for the latest version PDF URL
6. Otherwise → report failure with title/authors so the user can request via ILL

If only a title is given, resolve to a DOI first via Semantic Scholar search_paper_by_title (asta MCP) or Crossref.

Usage

python scripts/fetch.py <DOI> [options]
python scripts/fetch.py --batch <FILE|-> [options]
python scripts/fetch.py schema           # machine-readable self-description

Flags

Flag	Default	Description
doi	—	DOI to fetch (positional). Use - to read a single DOI from stdin
--batch FILE	—	File with one DOI per line for bulk download. Use - to read from stdin
--out DIR	pdfs	Output directory
--dry-run	off	Resolve sources without downloading; preview PDF URL and destination
--format	auto	json for agents, text for humans. Auto-detects: json when stdout is not a TTY, text when it is
--pretty	off	Pretty-print JSON with 2-space indent
--stream	off	Emit one NDJSON per line on stdout as each DOI resolves, then a summary line (batch mode)
--overwrite	off	Re-download even when destination file already exists
--idempotency-key KEY	—	Safe-retry key. Re-running with the same key replays the original envelope from <out>/.paper-fetch-idem/ without network I/O
--timeout SECONDS	30	HTTP timeout per request
--version	—	Print CLI + schema version and exit

Agent discovery: schema subcommand

python scripts/fetch.py schema

Emits a complete machine-readable description of the CLI on stdout (no network). Includes cli_version, schema_version, parameter types, exit codes, error codes, envelope shapes, and environment variables. Agents should read this once, cache it against schema_version, and re-read when the cached version drifts.

Output contract

stdout emits a single JSON envelope. Every envelope carries a meta slot.

Success (all DOIs resolved):

{
  "ok": true,
  "data": {
    "results": [
      {
        "doi": "10.1038/s41586-021-03819-2",
        "success": true,
        "source": "unpaywall",
        "pdf_url": "https://www.nature.com/articles/s41586-021-03819-2.pdf",
        "file": "pdfs/Jumper_2021_Highly_accurate_protein_structure_predic.pdf",
        "meta": {"title": "Highly accurate protein structure prediction with AlphaFold", "year": 2021, "author": "Jumper"},
        "sources_tried": ["unpaywall"]
      }
    
          
     
  
   
     
     
     
     
     
  

Partial (batch mode — some DOIs failed, exit code reflects the failure class):

{
  "ok": "partial",
  "data": {
    "results": [
      { "doi": "10.1038/s41586-021-03819-2", "success": true, "source": "unpaywall", ... },
      {
        "doi": "10.1234/nonexistent",
        "success": false,
        "source": null,
        "pdf_url": null,
        "file": null,
        "meta": {},
        "sources_tried":  
         
           
           
           
           
           
        
      
    
          
     
  
    ... 

The next slot is an array of suggested follow-up commands: re-invoking them retries only the failed subset. Combine with --idempotency-key to make the whole batch safely retriable without re-downloading the already-succeeded items.

Failure (bad arguments, exit code 3):

{
  "ok": false,
  "error": {
    "code": "validation_error",
    "message": "Provide a DOI or --batch file",
    "retryable": false
  },
  "meta": { ... }
}

Per-item skipped (destination already exists, no --overwrite):

{
  "doi": "10.1038/s41586-021-03819-2",
  "success": true,
  "source": "unpaywall",
  "pdf_url": "https://...",
  "file": "pdfs/Jumper_2021_...pdf",
  "skipped": true,
  "skip_reason": "file_exists",
  "sources_tried": ["unpaywall"]
}

Idempotency replay (re-run with the same --idempotency-key):

The cached envelope is returned verbatim, but meta.request_id and meta.latency_ms are re-stamped for the current call, and meta.replayed_from_idempotency_key is set. No network I/O occurs.

Stderr progress (NDJSON)

When --format json, stderr emits one JSON object per line for liveness:

{"event": "start",       "request_id": "req_...", "elapsed_ms": 2,    "doi": "10.1038/..."}
{"event": "source_try",  "request_id": "req_...", "elapsed_ms": 2,    "doi": "...", "source": "unpaywall"}
{"event": "source_hit",  "request_id": "req_...", "elapsed_ms": 2036, "doi": "...", "source": "unpaywall", "pdf_url": "..."}
{"event": "download_ok", "request_id": "req_...", "elapsed_ms": 4120, "doi": "...", "file": "..."}

Event types: start, source_try, source_hit, source_miss, source_skip, source_enrich, source_enrich_failed, download_ok, download_error, download_skip, dry_run, not_found, update_check_spawned. All events share request_id and elapsed_ms, letting an orchestrator correlate progress across stderr and the final stdout envelope.

source_enrich fires when Semantic Scholar is called purely to backfill missing author / title after another source already provided the PDF URL; its fields array lists exactly which fields were filled in. source_enrich_failed fires when that enrichment call fails — the Unpaywall PDF URL is still used and the filename falls back to unknown_<year>_….

When --format text, stderr emits human-readable prose.

Exit codes

Code	Meaning	Retryable class
0	All DOIs resolved / previewed	—
1	Unresolved — one or more DOIs had no OA copy; no transport failure	Not now (retry after retry_after_hours)
2	Reserved for auth errors (currently unused)	—
3	Validation error (bad arguments, missing input)	No
4	Transport error (network / download / IO failure)	Yes

The taxonomy lets an orchestrator route failures deterministically: exit 4 is worth retrying immediately, exit 1 is not, exit 3 is a bug in the caller.

Error codes in JSON

Code	Meaning	Retryable
validation_error	Bad arguments or empty input	No
not_found	No open-access PDF found	Yes (retry after retry_after_hours: 168)
download_network_error	Network failure during download	Yes
download_not_a_pdf	Response was not a PDF (HTML landing page)	No
download_host_not_allowed	PDF URL host not in allowlist	No
download_size_exceeded	Response exceeded 50 MB limit	Yes
download_io_error	Local filesystem write failed	Yes
internal_error	Unexpected error	No

Examples

# Single DOI (JSON output when piped; text when in a terminal)
python scripts/fetch.py 10.1038/s41586-020-2649-2

# Dry-run preview (resolve without downloading)
python scripts/fetch.py 10.1038/s41586-020-2649-2 --dry-run

# Force JSON (for agents even inside a terminal)
python scripts/fetch.py 10.1038/s41586-020-2649-2 --format json

# Human-readable with pretty colors in a pipeline
python scripts/fetch.py 10.1038/s41586-020-2649-2 --format text

# Batch download, safely retriable
python scripts/fetch.py --batch dois.txt --out ./papers \
    --idempotency-key monday-review-batch

# Pipe DOIs from another tool
zot -F ids.json query ... | jq -r '.[].doi' | python scripts/fetch.py --batch -

# Agent discovery
python scripts/fetch.py schema --pretty

# Streaming mode — one result per line as each DOI resolves
python scripts/fetch.py --batch dois.txt --stream

# Works without UNPAYWALL_EMAIL (skips Unpaywall, uses remaining 4 sources)
python scripts/fetch.py 10.1038/s41586-020-2649-2

Environment

Variable	Default	Purpose
UNPAYWALL_EMAIL	unset	Contact email for Unpaywall API. Optional but recommended. Without it, Unpaywall is skipped (remaining 4 sources still work).
PAPER_FETCH_ALLOWED_HOSTS	unset	Comma-separated extra hostnames to extend the download allowlist
PAPER_FETCH_NO_AUTO_UPDATE	unset	Set to any value to disable silent background self-update
PAPER_FETCH_UPDATE_INTERVAL	86400	Cooldown in seconds between update checks

Notes

• Auth is delegated. The agent never runs a login subcommand. The human or the orchestrator sets UNPAYWALL_EMAIL in the environment; the agent inherits it. Missing email degrades gracefully to the remaining 4 sources.
• Trust is directional. CLI arguments are validated once at the entry point. The host allowlist and 50 MB size cap are enforced in the environment layer, not at the agent's request. An agent cannot loosen safety by passing a flag — only by the operator setting PAPER_FETCH_ALLOWED_HOSTS.
• Downloads are naturally idempotent. Re-running against the same --out skips files that already exist (deterministic filename: {first_author}_{year}_{short_title}.pdf). Pair with --idempotency-key to also replay the exact envelope without any network I/O.
• Never attempts to bypass paywalls. If no OA copy exists, the skill reports failure honestly — do not suggest Sci-Hub or similar.
• Default output directory: ./pdfs/.

Auto-update

When installed via git clone, the skill keeps itself in sync with upstream automatically. On each invocation, fetch.py spawns a detached background git pull --ff-only in the skill directory:

• Non-blocking — the current invocation is not delayed; the pull runs in a new session and is fully detached
• Silent — all git output goes to /dev/null, the stdout envelope is never polluted
• Throttled — at most once every 24 hours (stamped via .git/.paper-fetch-last-update)
• Safe — --ff-only refuses to merge if you have local edits; conflicts never happen
• Observable — when a pull is spawned, stderr emits {"event": "update_check_spawned", ...} (JSON mode) or a prose notice (text mode)
• Convergence — updates apply on the next invocation, not the current one (because the pull is backgrounded)

Force an immediate check with rm <skill_dir>/.git/.paper-fetch-last-update.