- Exploited
- No known exploitation
- Fix
- Fixed in 3.6.0
- Who is exposed
- Not reported
Do first: Upgrade to a fixed version: 3.6.0.
- Last modified
- A 0-100 score for how much a CVE matters across the industry: exploitation, exposure, how widely the software is used, severity and whether a fix exists.What do these mean? →
- 16/100 · provisional
- Whether working attack code is public (Exploit-DB, Metasploit, Nuclei or GitHub). Public code makes attacks easier.What do these mean? →
- None on Exploit-DB, Metasploit or GitHub
Summary
Earlier releases of rmcp's Streamable HTTP server transport let any client that can reach the endpoint exhaust server memory by sending `tools/call` requests with many distinct, unknown tool names, because the transport caches each name without limit, even when the request is rejected with HTTP 400. Only servers that handle newer protocol revisions are affected, and servers without an authentication layer in front of the endpoint are exposed because rmcp does not authenticate requests itself.
Written by anthropic/claude-haiku-5.5 from the sources below. Verify against them.
Read the original description
From 3.0.0 until 3.6.0, StreamableHttpService cached the result of ServerHandler::get_tool in a process-wide map keyed by the tool name of incoming tools/call requests. The map had no size limit and no eviction, and it also cached unknown tool names.
An entry was added while the request headers were validated, before the request could be rejected, so requests answered with HTTP 400 still grew the map. A client that can reach the endpoint can send tools/call requests with distinct tool names, each up to the default 4 MiB body limit, until the server runs out of memory. rmcp does not authenticate requests itself, so servers without an authentication layer in front of the endpoint are exposed to unauthenticated clients.
Only servers that use the Streamable HTTP server transport and receive requests at protocol version 2026-07-28 or later are affected.
The flaw was corrected in 3.6.0 by removing the cache. The transport now calls ServerHandler::get_tool on each request.
If you cannot upgrade, put authentication in front of the endpoint, or reject tools/call requests for unknown tool names before they reach StreamableHttpService.
cve.now Index
- Exploitation?/40Unknown. No CISA SSVC exploitation rating yet
- Exposure?/25Unknown. No advisory context yet
- Reach16/20rmcp: 2.1M downloads a week
- Severity?/10Unknown. No CVSS score or vendor rating
- No patch0/5No signal
What would change the score
- Would score about 71 if added to CISA KEV
- Would score about 66 if CISA's SSVC rates exploitation active
- Would score about 38 if advisories confirm default setups are exposed
Impact
- Widely used
rmcp has 2.1m weekly downloads.Source: crates.io
What to do
Upgrade to a fixed version: 3.6.0.
Timeline
- Advisory published20 days
- Record last updated
Affected versions
- rmcpcrates.io
From 3.0.0 · fixed in 3.6.0
Fixed in 3.6.0 (released Oct 20, 2026, the day of disclosure).
The latest release, 3.9.0, is not affected.
Evidence 3
- Advisory
- Repository
- Link
Corrections
None so far.
Something wrong on this page? Report an error. Every fix is listed here.
Questions about RUSTSEC-2026-0461
Is RUSTSEC-2026-0461 being exploited?
It is not in CISA's Known Exploited Vulnerabilities catalog, and CISA's SSVC assessment does not mark exploitation as active. That does not mean it has not been attacked: vendors and researchers can report attacks before CISA lists them. No public exploit code is tracked.Source: cve.now feed data
Which versions are affected by RUSTSEC-2026-0461?
rmcp (crates.io): from 3.0.0 before 3.6.0.Source: cve.now feed data
How do I fix RUSTSEC-2026-0461?
Upgrade to a fixed version: 3.6.0.Source: cve.now feed data