Claidex MCP server
com.claidex/failure-intelligenceclaidex.comstreamable-httpv1.0.0Scanned
Biomedical failure intelligence platform for terminated trials and translational analysis.
24 tools: 6 change things and 18 only read.
- Reads private data · 4
- Reads text others write · 3
- Can send data out · 1
- No findings
Sign-in required, no OAuth metadata. Remote posture
With Anlyon
6 tools here change things. Govern them.
- Anlyon lists the tools and adopts none.
- Each tool you adopt gets approvals and receipts.
- Your agent never holds the key.
A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.
What the Claidex MCP server can do
The Claidex MCP server is published by claidex.com, a domain the MCP Registry verified. It has 24 tools. It can create or change data (6 tools: claidex_research_prompt, consult_specialist_model, rank_documents_by_embedding, rerank_documents and 2 more), but cannot delete data. 18 tools only read.
4 tools read private data (query_openfda, search_claims, search_preprint_flags and 1 more); 3 tools bring in text other people can write (fetch_research_url, search_claims and web_search); 1 tool can send data out of your system (fetch_research_url). On its own it has all three parts of a prompt-injection exfiltration path: private data, text others write, and a way out.
None of the fixed checks matched anything in its tool definitions. That is not a safety guarantee.
Last scanned Nov 10, 2026, version 1.0.0.
Tools
Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.
24 tools
Changes things 6Claude Code: ask
claidex_research_promptCompose a rigorous, reusable investigation prompt that tells an MCP client which Claidex tools and resources to use.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
Description, as the server wrote it Compose a rigorous, reusable investigation prompt that tells an MCP client which Claidex tools and resources to use.
consult_specialist_modelAsk a specialist Gateway model for code, database/tool-calling, statistics, literature, preprint, or mechanistic synthesis support.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
Description, as the server wrote it Ask a specialist Gateway model for code, database/tool-calling, statistics, literature, preprint, or mechanistic synthesis support. Use for hard subproblems, second opinions, code generation/review, or complex synthesis.
rank_documents_by_embeddingEmbed a query and candidate documents, then rank documents by cosine similarity.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
Description, as the server wrote it Embed a query and candidate documents, then rank documents by cosine similarity. Use for semantic matching, retrieval checks, clustering triage, and lightweight RAG over user-provided passages.
rerank_documentsRerank candidate documents/passages against a query using a dedicated reranking model.No findings
Why it is in this group
- Nothing in its name or description says what it does, so it is counted as a write.
Description, as the server wrote it Rerank candidate documents/passages against a query using a dedicated reranking model. Use after retrieval/search when exact relevance ordering matters.
run_computationEvaluate deterministic mathematical expressions with optional variables.No findings
Why it is in this group
- The name "run_computation" says "run": a word for a change.
Description, as the server wrote it Evaluate deterministic mathematical expressions with optional variables. Use for algebra, matrices, vectors, transformations, descriptive numerical checks, and reproducible calculations. Does not execute arbitrary JavaScript or install packages.
run_statistical_analysisRun a statistical analysis computation.No findings
Why it is in this group
- The name "run_statistical_analysis" says "run": a word for a change.
Description, as the server wrote it Run a statistical analysis computation. Supports: Wilson confidence intervals, binomial proportions, phase transition success rates, power calculations, Bayesian posterior estimates, descriptive statistics, and two-proportion comparisons. Returns computed values, interpretation, and LaTeX formula.
Only reads 18Show 18 read-only tools
claidex_capabilitiesReturn the Claidex MCP feature map, configured storage/model providers, safety controls, resources, prompts, and tool counts.No findings
Why it is in this group
- The description's first verb is "return": a word for reading.
Description, as the server wrote it Return the Claidex MCP feature map, configured storage/model providers, safety controls, resources, prompts, and tool counts.
claidex_claim_risk_matrixSearch claims and summarize MRS bands, trial phases, failure archetypes, and high-risk programs for rapid diligence.No findings
Why it is in this group
- The description's first verb is "search": a word for reading.
Description, as the server wrote it Search claims and summarize MRS bands, trial phases, failure archetypes, and high-risk programs for rapid diligence.
claidex_health_snapshotReturn Sentinel metrics, coverage, graph counts, recent automation runs, MRS bands, and preprint-watch summaries.No findings
Why it is in this group
- The name "claidex_health_snapshot" says "snapshot": a word for reading.
Description, as the server wrote it Return Sentinel metrics, coverage, graph counts, recent automation runs, MRS bands, and preprint-watch summaries.
claidex_tool_catalogReturn the complete MCP tool catalog, optionally including full input schemas and filtering native versus agent-backed tools.No findings
Why it is in this group
- The description's first verb is "return": a word for reading.
Description, as the server wrote it Return the complete MCP tool catalog, optionally including full input schemas and filtering native versus agent-backed tools.
fetchFetch a specific result returned by search.No findings
Why it is in this group
- The name "fetch" says "fetch": a word for reading.
Description, as the server wrote it Fetch a specific result returned by search. This compatibility tool supports OpenAI and ChatGPT remote MCP retrieval flows.
fetch_research_urlFetch a public HTTPS research URL, JSON API response, package metadata endpoint, documentation page, or dataset preview.Reads text others writeCan send data outNo findings
Why it is in this group
- The name "fetch_research_url" says "fetch": a word for reading.
What it can reach
- untrusted-inputThe name "fetch_research_url" says "fetch" and "url": it reads web pages, which anyone can write.
- untrusted-inputThe description's first sentence says "fetch" and "http": it reads web pages, which anyone can write.
- untrusted-inputThe input takes a URL (input.url), and the tool reads it: it reads web pages, which anyone can write.
- outboundThe input takes a URL (input.url): a request to a URL can carry data to whoever runs it.
Claude Code: ask, because it is tagged outbound.
Description, as the server wrote it Fetch a public HTTPS research URL, JSON API response, package metadata endpoint, documentation page, or dataset preview. Blocks private/internal hosts and truncates large responses.
get_claim_contentRetrieve the complete text content of a specific Claidex Claim by its slug.No findings
Why it is in this group
- The name "get_claim_content" says "get": a word for reading.
Description, as the server wrote it Retrieve the complete text content of a specific Claidex Claim by its slug. Use this when you need to read the full post-mortem analysis, including hypothesis, failure mechanism, and prevention analysis.
query_chemblLook up a drug or compound in ChEMBL.No findings
Why it is in this group
- The name "query_chembl" says "query": a word for reading.
Description, as the server wrote it Look up a drug or compound in ChEMBL. Returns mechanism of action, primary target, ChEMBL ID, max development phase, molecular type, and synonyms.
query_clinicaltrialsSearch ClinicalTrials.gov for clinical trials.No findings
Why it is in this group
- The name "query_clinicaltrials" says "query": a word for reading.
Description, as the server wrote it Search ClinicalTrials.gov for clinical trials. Can search by drug name, condition, target, or NCT ID. Returns trial status, phase, sponsor, primary endpoints, and why_stopped if terminated.
query_failure_graphQuery the Claidex failure graph for a specific gene target.No findings
Why it is in this group
- The name "query_failure_graph" says "query": a word for reading.
Description, as the server wrote it Query the Claidex failure graph for a specific gene target. Returns the Mechanism Risk Score (MRS), failure counts by phase and archetype, and all related claim slugs.
query_openfdaQuery the FDA FAERS database for adverse event signals for a drug.Reads private dataNo findings
Why it is in this group
- The name "query_openfda" says "query": a word for reading.
What it can reach
- private-dataThe description's first sentence says "query" and "database": it reads databases.
Description, as the server wrote it Query the FDA FAERS database for adverse event signals for a drug. Returns top adverse event terms, serious case count, and death count.
query_opentargetsQuery Open Targets for target-disease association evidence.No findings
Why it is in this group
- The name "query_opentargets" says "query": a word for reading.
Description, as the server wrote it Query Open Targets for target-disease association evidence. Returns genetic evidence score, clinical evidence, known drugs, tractability, and safety liabilities for a gene-disease pair.
query_package_registryLook up package metadata from npm or PyPI before recommending code, analysis packages, API clients, or dataset tooling.No findings
Why it is in this group
- The name "query_package_registry" says "query": a word for reading.
Description, as the server wrote it Look up package metadata from npm or PyPI before recommending code, analysis packages, API clients, or dataset tooling.
searchSearch Claidex failure claims and preprint flags.No findings
Why it is in this group
- The name "search" says "search": a word for reading.
Description, as the server wrote it Search Claidex failure claims and preprint flags. This compatibility tool returns citation-ready results for OpenAI and ChatGPT remote MCP flows.
search_claimsSearch the Claidex failure database for post-mortems matching a drug name, gene target, disease, failure archetype, or sponsor.Reads private dataReads text others writeNo findings
Why it is in this group
- The name "search_claims" says "search": a word for reading.
What it can reach
- private-dataThe description's first sentence says "search" and "database": it reads databases.
- untrusted-inputThe description's first sentence says "search" and "post": it reads posts, which other people can write.
Description, as the server wrote it Search the Claidex failure database for post-mortems matching a drug name, gene target, disease, failure archetype, or sponsor. Returns matching claims with title, slug, drug, target, disease, phase, failure archetype, Open Targets score, and MRS score.
search_preprint_flagsSearch the Claidex Preprint Watch database for bioRxiv preprints flagged as contradicting the clinical failure record.Reads private dataNo findings
Why it is in this group
- The name "search_preprint_flags" says "search": a word for reading.
What it can reach
- private-dataThe description's first sentence says "search" and "database": it reads databases.
Description, as the server wrote it Search the Claidex Preprint Watch database for bioRxiv preprints flagged as contradicting the clinical failure record. Search by target gene, disease, or preprint title.
search_pubmedSearch PubMed for biomedical literature.No findings
Why it is in this group
- The name "search_pubmed" says "search": a word for reading.
Description, as the server wrote it Search PubMed for biomedical literature. Returns title, authors, journal, year, DOI, abstract, and PMID for each result.
web_searchSearch the web for recent news, investor commentary, analyst notes, press releases, and conference coverage about drugs, clinical trials, or biomedical mechanisms.Reads private dataReads text others writeNo findings
Why it is in this group
- The name "web_search" says "search": a word for reading.
What it can reach
- private-dataThe description's first sentence says "search" and "note": it reads notes.
- untrusted-inputThe name "web_search" says "search" and "web": it reads web pages, which anyone can write.
- untrusted-inputThe description's first sentence says "search" and "web": it reads web pages, which anyone can write.
Description, as the server wrote it Search the web for recent news, investor commentary, analyst notes, press releases, and conference coverage about drugs, clinical trials, or biomedical mechanisms.
This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.
Exfil surface
Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.
12 paths in this server alone from private data, through text someone else wrote, to a way out. Denying 1 outbound tool breaks every one.
- Reads private data4
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write3
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out1
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
{
"permissions": {
"deny": [
"mcp__failure-intelligence__fetch_research_url"
]
}
}Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.
Remote posture
What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.
Looked at Nov 20, 2026, 18:35 UTC, unchanged since Nov 13, 2026. No credentials were sent and no tool was called. What is looked at.
- Reachable
- yes
- HTTP status of an anonymous initialize
- 404
- tools/list served without credentials
- no
- Tools that change things, listed without credentials
- 0
- Protocol version
- not observed
- WWW-Authenticate on 401
- not observed
- resource_metadata in WWW-Authenticate
- not observed
- RFC 9728 protected resource metadata
- not observed
- Authorization servers
- not observed
- Legacy HTTP+SSE transport
- no
- Dynamic client registration
- not observed
- Client ID metadata documents
- not observed
The probe stopped: initialize answered HTTP 404.
Posture changes
- Nov 13, 2026
- HTTP status of an anonymous initialize: 200 to 404
- tools/list served without credentials: yes to no
- Tools that change things, listed without credentials: 6 to 0
- Protocol version: 2025-06-18 to none
Findings
Matches of fixed checks, with the text they matched.
No check matched anything in these tool definitions. That is not a safety guarantee.
History
Changes to its tool definitions and to its remote endpoint, newest first.
PostureRemote posture changed. HTTP status of an anonymous initialize: 200 -> 404; tools/list served without credentials: yes -> no; Tools that change things, listed without credentials: 6 -> 0 (and 1 more).View diff
- HTTP status of an anonymous initialize: 200 to 404
- tools/list served without credentials: yes to no
- Tools that change things, listed without credentials: 6 to 0
- Protocol version: 2025-06-18 to none
Install
Pinned to version 1.0.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http failure-intelligence https://app.claidex.com/api/mcp --header 'Authorization: <your-authorization>'claude mcp add --transport http failure-intelligence https://app.claidex.com/api/mcp --header 'Authorization: <your-authorization>'Run in a terminal.
Claude Code permissions
Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.
{
"permissions": {
"allow": [
"mcp__failure-intelligence__claidex_capabilities",
"mcp__failure-intelligence__claidex_claim_risk_matrix",
"mcp__failure-intelligence__claidex_health_snapshot",
"mcp__failure-intelligence__claidex_tool_catalog",
"mcp__failure-intelligence__fetch",
"mcp__failure-intelligence__get_claim_content",
"mcp__failure-intelligence__query_chembl",
"mcp__failure-intelligence__query_clinicaltrials",
"mcp__failure-intelligence__query_failure_graph",
"mcp__failure-intelligence__query_openfda",
"mcp__failure-intelligence__query_opentargets",
"mcp__failure-intelligence__query_package_registry",
"mcp__failure-intelligence__search",
"mcp__failure-intelligence__search_claims",
"mcp__failure-intelligence__search_preprint_flags",
"mcp__failure-intelligence__search_pubmed",
"mcp__failure-intelligence__web_search"
],
"ask": [
"mcp__failure-intelligence__claidex_research_prompt",
"mcp__failure-intelligence__consult_specialist_model",
"mcp__failure-intelligence__fetch_research_url",
"mcp__failure-intelligence__rank_documents_by_embedding",
"mcp__failure-intelligence__rerank_documents",
"mcp__failure-intelligence__run_computation",
"mcp__failure-intelligence__run_statistical_analysis"
],
"deny": []
}
}Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as failure-intelligence. Review it before you use it.
Capability tags behind 5 rules
- mcp__failure-intelligence__fetch_research_url untrusted-input, outbound
- mcp__failure-intelligence__query_openfda private-data
- mcp__failure-intelligence__search_claims private-data, untrusted-input
- mcp__failure-intelligence__search_preprint_flags private-data
- mcp__failure-intelligence__web_search private-data, untrusted-input
Pin
Keep the decision: fail CI when the tools change, follow every change, or show a badge.
Pin in CI
Fail a build when the tools you reviewed are not the tools the server serves today.
curl -s "https://radar.anlyon.com/api/v1/servers/com.claidex/failure-intelligence/verify?hash=212f388a7a34f408f8193546259650969d25a6bd08f6786dcdd122e6c963ed44" | jq -e '.changed == false'Pinned hash 212f388a7a34, from Nov 10, 2026, 12:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/com.claidex/failure-intelligence/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/com.claidex/failure-intelligence)Source: MCP Registry · repository. Synced Oct 21, 2026, 17:52 UTC.
Claidex MCP server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the Claidex MCP server safe?
- No scan can promise that. What its definitions show: 24 tools, 6 of which can change, send or delete something, and 0 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.
Can the Claidex MCP server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the Claidex MCP server do things that cannot be undone?
- No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.
What tools does the Claidex MCP server have?
- 24 tools: 6 write, 18 read. claidex_research_prompt, consult_specialist_model, rank_documents_by_embedding, rerank_documents, run_computation, run_statistical_analysis, claidex_capabilities, claidex_claim_risk_matrix and 16 more.
Does the Claidex MCP server need sign-in or OAuth?
- Its remote endpoint did not answer the last anonymous look.
Does the Claidex MCP server send data to outside services?
- 1 tool can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): fetch_research_url.
Has the Claidex MCP server changed its tools?
- Not since Anlyon Radar started tracking it. Last scanned Nov 10, 2026.