osv-advisory-mcp-server
io.github.cyanheads/osv-advisory-mcp-servercyanheadsstreamable-httpv0.6.0Scanned
Query OSV.dev for package vulnerabilities and batch-audit dependency lists via MCP.
5 tools: 5 only read.
- Reads private data · 1
- Reads text others write · 4
- Can send data out · 0
- No findings
Tools listed without sign-in. Remote posture
With Anlyon
Decide which of this server's tools your agent can use.
- 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 osv-advisory-mcp-server can do
The osv-advisory-mcp-server is published by the GitHub account cyanheads. It has 5 tools. Every one of them only reads: going by their definitions, none of them creates, changes, sends or deletes anything.
1 tool reads private data (osv_query_commit); 4 tools bring in text other people can write (osv_query_package, osv_get_vulnerability, osv_query_batch and 1 more).
None of the fixed checks matched anything in its tool definitions. That is not a safety guarantee.
Its remote endpoint lists its tools without sign-in, and its tools changed 4 times since Anlyon Radar first saw them.
Last scanned Nov 20, 2026, version 0.6.0.
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.
5 tools
Only reads 5Show 5 read-only tools
osv_get_vulnerabilityFetch the full advisory record for an OSV vulnerability ID.Reads text others writeNo findings
Why it is in this group
- The name "osv_get_vulnerability" says "get": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Fetch the full advisory record for an OSV vulnerability ID. Returns the complete record: summary, full details text, CVE aliases, all affected packages and version ranges, fix versions, CVSS severity vectors, CWE weakness IDs, and references. Use when osv_query_package or osv_query_batch returns a vuln ID and you need the full advisory context — affected packages or remediation guidance. For overflow, repeat the original input with page: overflow.page. Continue the same path with offset: page.nextOffset and revision: page.revision until nextOffset is null. Retrieve each entries[].deferred.path separately at offset zero without a revision. Do not pass the output page object as input. Every page is a fresh lookup; restart a changed node without a revision. Transmission overflow and upstream truncation are separate; neither confirms clean.
Annotations the server declares
- readOnlyHint: true
- idempotentHint: true
- openWorldHint: true
osv_list_ecosystemsReturn the supported ecosystem base names for osv_query_package and osv_query_batch: every ecosystem the OSV schema names that OSV.dev accepts at query time, plus GIT, as verified …No findings
Why it is in this group
- The name "osv_list_ecosystems" says "list": a word for reading.
Description, as the server wrote it Return the supported ecosystem base names for osv_query_package and osv_query_batch: every ecosystem the OSV schema names that OSV.dev accepts at query time, plus GIT, as verified on 2026-10-31. Known base names are case-insensitive. Preserve suffix spelling. Alpine needs a release branch, such as Alpine:v3.17. The list is static and may lag ecosystems added after that date.
Annotations the server declares
- readOnlyHint: true
- idempotentHint: true
- openWorldHint: false
osv_query_batchQuery vulnerabilities for multiple packages in one call — the primary tool for dependency audits, SBOM scanning, and lockfile triage.Reads text others writeNo findings
Why it is in this group
- The name "osv_query_batch" says "query": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Query vulnerabilities for multiple packages in one call — the primary tool for dependency audits, SBOM scanning, and lockfile triage. Submit up to 5000 mixed {name, ecosystem, version} tuples and {purl, version?} identifiers, with exactly one embedded or separate version per PURL. The first 1000 entries are processed. Follow summary.nextInputOffset to submit the remaining slice. Pass a row's nextCursor as cursor to osv_query_package with that row's identifier. Each entry in the response corresponds positionally to the input. Each finding includes CVE aliases for chaining to nist-nvd-mcp-server for CVSS scoring. Large replies retain the complete summary. For overflow, repeat the original packages array with page: overflow.page. Continue the same path with offset: page.nextOffset and revision: page.revision until nextOffset is null. Retrieve each entries[].deferred.path separately at offset zero without a revision. Do not pass the output page object as input. /results/17 selects original input row 17, even when earlier rows failed. Pages are fresh lookups and do not recompute the initial summary; restart a changed node without a revision. Transmission overflow and upstream truncated are separate; neither confirms clean.
Annotations the server declares
- readOnlyHint: true
- idempotentHint: true
- openWorldHint: true
osv_query_commitQuery known vulnerabilities for one full Git commit hash (40 hexadecimal SHA-1 or 64 hexadecimal SHA-256 characters, case-insensitive).Reads private dataReads text others writeNo findings
Why it is in this group
- The name "osv_query_commit" says "query": a word for reading.
What it can reach
- private-dataThe name "osv_query_commit" says "query" and "commit": it reads repositories.
- private-dataThe description's first sentence says "query" and "commit": it reads repositories.
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Query known vulnerabilities for one full Git commit hash (40 hexadecimal SHA-1 or 64 hexadecimal SHA-256 characters, case-insensitive). Abbreviated hashes are rejected. Returns advisory summaries, record-wide severity and GIT ranges with repositories and ordered fix events; use osv_get_vulnerability for full records. No package-specific fixedVersions are inferred. For overflow, repeat the original input with page: overflow.page. Continue the same path with offset: page.nextOffset and revision: page.revision until nextOffset is null. Retrieve entries[].deferred.path separately at offset zero without a revision. Do not pass the output page object as input. Pages are fresh lookups; restart a changed node without a revision. Transmission overflow and upstream truncated are separate; neither confirms clean.
Annotations the server declares
- readOnlyHint: true
- idempotentHint: true
- openWorldHint: true
osv_query_packageQuery known vulnerabilities for a single package version using {name, ecosystem, version} or {purl, version?}; supply exactly one embedded or separate version.Reads text others writeNo findings
Why it is in this group
- The name "osv_query_package" says "query": a word for reading.
What it can reach
- untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
Description, as the server wrote it Query known vulnerabilities for a single package version using {name, ecosystem, version} or {purl, version?}; supply exactly one embedded or separate version. Release-qualified Ubuntu names are resolved to the source packages OSV indexes. Returns all matching OSV advisories with severity (CVSS vectors), CVE aliases, affected version ranges, and the fixed versions listed for the queried package. Find canonical ecosystem names and the Alpine release format in osv_list_ecosystems. Known base names are case-insensitive. Preserve suffix spelling. For overflow, repeat the original input with page: overflow.page. Continue the same path with offset: page.nextOffset and revision: page.revision until nextOffset is null. Retrieve each entries[].deferred.path separately at offset zero without a revision. Do not pass the output page object as input. Set page.advisoryId to an announced ID to fetch its fields with fixes and severity for the queried package; retain advisoryId on subsequent pages. Pages are fresh lookups; restart a changed node without a revision. Transmission overflow and upstream truncated are separate; neither confirms clean.
Annotations the server declares
- readOnlyHint: true
- idempotentHint: true
- openWorldHint: true
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.
No private-data, untrusted-input, outbound path in this server alone: no tool is tagged outbound.
- Reads private data1
- Reads files, mail, messages, databases, secrets, calendars or repositories.
- Reads text others write4
- Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
- Can send data out0
- Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or 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, 19:33 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.
- Reachable
- yes
- HTTP status of an anonymous initialize
- 200
- tools/list served without credentials
- yes
- Tools that change things, listed without credentials
- 0
- Protocol version
- 2025-06-18
- 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
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.
Tools2 changed.~2 changedView diff
- Changed:osv_query_batchOnly readsinputSchema
- Changed:osv_query_packageOnly readsdescription
Tools5 changed.~5 changedView diff
- Changed:osv_get_vulnerabilityOnly readsannotations
- Changed:osv_list_ecosystemsOnly readsdescription
- Changed:osv_query_batchOnly readsinputSchemaannotations
- Changed:osv_query_commitOnly readsannotations
- Changed:osv_query_packageOnly readsdescriptioninputSchemaannotations
Tools3 changed.~3 changedView diff
- Changed:osv_query_batchOnly readsdescriptioninputSchema
- Changed:osv_query_commitOnly readsinputSchema
- Changed:osv_query_packageOnly readsinputSchema
Tools3 changed.~3 changedView diff
- Changed:osv_list_ecosystemsOnly readsdescription
- Changed:osv_query_batchOnly readsinputSchema
- Changed:osv_query_packageOnly readsinputSchema
Install
Pinned to version 0.6.0. Values in angle brackets are yours to fill in.
claude mcp add --transport http osv-advisory https://osv-advisory.caseyjhand.com/mcpclaude mcp add --transport http osv-advisory https://osv-advisory.caseyjhand.com/mcpRun 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__osv-advisory__osv_list_ecosystems",
"mcp__osv-advisory__osv_query_package",
"mcp__osv-advisory__osv_get_vulnerability",
"mcp__osv-advisory__osv_query_batch",
"mcp__osv-advisory__osv_query_commit"
],
"ask": [],
"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 osv-advisory. Review it before you use it.
Capability tags behind 4 rules
- mcp__osv-advisory__osv_query_package untrusted-input
- mcp__osv-advisory__osv_get_vulnerability untrusted-input
- mcp__osv-advisory__osv_query_batch untrusted-input
- mcp__osv-advisory__osv_query_commit 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/io.github.cyanheads/osv-advisory-mcp-server/verify?hash=80ee6219d8263eb29ef78087d7236b85095db0cbc40175d334ecdba79fa7fb72" | jq -e '.changed == false'Pinned hash 80ee6219d826, from Nov 20, 2026, 23:18 UTC.
Watch
Get told when its tools change: a new tool, a new description, a new class.
https://radar.anlyon.com/servers/io.github.cyanheads/osv-advisory-mcp-server/changes.xmlAn Atom feed. Works in any feed reader or Slack's RSS app.
Badge
[](https://radar.anlyon.com/servers/io.github.cyanheads/osv-advisory-mcp-server)Source: MCP Registry · repository. Synced Nov 20, 2026, 23:17 UTC.
osv-advisory-mcp-server: questions and answers
Answered from the scan above, not from the publisher's description.
Is the osv-advisory-mcp-server safe?
- No scan can promise that. What its definitions show: 5 tools, all of which only read, and 0 findings from fixed checks. Read-only tools can still pass private data to a model.
Can the osv-advisory-mcp-server delete data?
- No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.
Can the osv-advisory-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 osv-advisory-mcp-server have?
- 5 tools: 5 read. osv_list_ecosystems, osv_query_package, osv_get_vulnerability, osv_query_batch and osv_query_commit.
Does the osv-advisory-mcp-server need sign-in or OAuth?
- Its remote endpoint listed its tools to an anonymous client, without sign-in.
Does the osv-advisory-mcp-server send data to outside services?
- None of its tools sends, posts or publishes, or takes a recipient, channel or URL.
Has the osv-advisory-mcp-server changed its tools?
- Yes, 4 times since Anlyon Radar started tracking it, most recently on Nov 20, 2026. Every change is on this page and in its Atom feed.