Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add cyanheads/pubmed-mcp-server --skill pubmed-guidegit clone --depth 1 https://github.com/cyanheads/pubmed-mcp-serverWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/cyanheads/pubmed-mcp-server/pubmed-guide)<a href="https://agentmods.dev/skills/cyanheads/pubmed-mcp-server/pubmed-guide"><img src="https://agentmods.dev/badge/skills/cyanheads/pubmed-mcp-server/pubmed-guide/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/cyanheads/pubmed-mcp-server/pubmed-guide"><img src="https://agentmods.dev/badge/skills/cyanheads/pubmed-mcp-server/pubmed-guide.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00082 | $0.02486 |
| Opus 5.5 | $0.00033 | $0.00994 |
| Sonnet 5.5 | $0.00016 | $0.00497 |
| Haiku 4.5 | $0.00008 | $0.00249 |
Grade A, and why
pubmed-guide scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Scope
Biomedical literature: PubMed records (MEDLINE, PubMed Central, NCBI Bookshelf), PMC full text, Europe PMC (adds preprints, patents, Agricola), and NIH iCite citation metrics. Keyless. NCBI_API_KEY raises NCBI's rate ceiling, UNPAYWALL_EMAIL adds an Unpaywall full-text tier, and a deployment that sets EUROPEPMC_ENABLED=false registers neither Europe PMC tool.
- Full text only where an open copy exists: PMC, Europe PMC's open-access subset, or, with
UNPAYWALL_EMAILset, a copy Unpaywall finds. A subscription article outside all three comes back inunavailable. - iCite metrics are keyed by PMID, so a work PubMed doesn't index has none here.
- Trial registrations appear only as accessions in
pubmed_fetch_articlesdataBanks; the registry record itself is a ClinicalTrials.gov server's job. Literature outside biomedicine belongs to an OpenAlex or Crossref server.
Identifiers
| ID | Form | Comes from |
|---|---|---|
| PMID | digits | pubmed_search_articles, pubmed_lookup_citation, a MED hit of pubmed_europepmc_search (its epmcId is the PMID) |
| PMCID | PMC + digits |
pmcId on pubmed_fetch_articles records and search summaries; pubmed_convert_ids |
| DOI | 10.…/… |
doi on pubmed_fetch_articles records and Europe PMC hits |
Europe PMC source + epmcId |
PPR1306463 |
pubmed_europepmc_search hits only; preprints and patents often carry no PMID or DOI |
| MeSH descriptor | its preferred name | name on pubmed_lookup_mesh results |
| ORCID | 0000-0002-5446-9969 |
orcid on pubmed_fetch_articles authors, when the record carries one |
Crosswalks:
- PMID → DOI or PMCID:
pubmed_fetch_articles, for any PubMed record. - DOI → PMID:
pubmed_convert_idsanswers only for articles in PMC;10.1056/NEJMoa1911303(not in PMC) fails there. Search10.1056/NEJMoa1911303[doi]withpubmed_search_articles(returns 31535829), orDOI:"10.1056/NEJMoa1911303"withpubmed_europepmc_search, whose hit carries the PMID. - ORCID → an author's papers:
0000-0002-5446-9969[auid]inpubmed_search_articlesreturned 46 records. It matches only records whose author entry carries that ORCID, however the name is spelled.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 80 lines · 82 tokens per session scan A 8a0fd124e755
pubmed-guide is a skill published in the GitHub repository cyanheads/pubmed-mcp-server (167 stars, last pushed 3d ago), licensed Apache-2.0. It adds 82 tokens to every session and 2,486 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-11-04.
pubmed-guide compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| pubmed-guidecyanheads/pubmed-mcp-server | Skill | 167 | 3d ago | Apache-2.0 | A | 82 | Claude Code |
| mitiq-error-mitigationxi-zhao/OpenQuantum | Skill | 71 | yesterday | MIT | A | 48 | Claude Code, Codex |
| quantum-ground-statexi-zhao/OpenQuantum | Skill | 71 | yesterday | MIT | A | 103 | Codex |
| quantum-amplitude-estimationxi-zhao/OpenQuantum | Skill | 71 | yesterday | MIT | A | 63 | Claude Code, Codex |
| quantum-groverxi-zhao/OpenQuantum | Skill | 71 | yesterday | MIT | A | 62 | Claude Code, Codex |
| quantum-hidden-shiftxi-zhao/OpenQuantum | Skill | 71 | yesterday | MIT | A | 47 | Claude Code, Codex |
| quantum-lcuxi-zhao/OpenQuantum | Skill | 71 | yesterday | MIT | A | 61 | Claude Code, Codex |