Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessSend HTTP request with crafted Markdown body
-
DeliverySubmit 1000+ consecutive asterisk characters surrounding text
-
ExploitInline parser builds unbounded nested emphasis token tree
-
ExecutionRecursive HTMLRenderer.render_token() exceeds Python recursion limit
-
PersistUnhandled RecursionError crashes worker process
-
ImpactService endpoint unavailable
Vulnerability Assessment
| Weakness |
CWE
ASVS 5.0
V15 - Secure Coding and Architecture Recursion depth limits
|
| Exploitation Conditions | The target application must accept user-supplied content and pass it to Mistune 3.3.0-3.3.2 for parsing and HTML rendering - this is the sole prerequisite. No authentication is required when the Markdown input endpoint is publicly accessible (CVSS PR:N confirmed). No non-default Mistune configuration or plugin is needed; the vulnerability is present in the default InlineParser with emphasis parsing enabled, which is the out-of-box state. The primary limiting factor is deployment pattern: applications that render Markdown only from trusted internal or author-controlled sources (static site generators, internal wikis with admin-only authorship) are not reachable by external attackers. Applications with public comment fields, issue trackers, API endpoints accepting Markdown bodies, or documentation platforms with user submissions are directly exploitable. |
| Impact | The CVSS 7.5 vector (AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H) is well-supported by the technical evidence: any network-reachable endpoint that pipes user input into Mistune 3.3.0-3.3.2 is exploitable with a single, trivially crafted HTTP request. Attack complexity is genuinely low - the fix commit's own test case is a 2001-character string, requiring no reverse engineering or environment-specific knowledge. Impact is bounded strictly to availability; there is no confidentiality or integrity exposure. No EPSS score is available in the provided intelligence, so exploitation probability cannot be quantified statistically; absence from CISA KEV indicates no confirmed active exploitation at time of analysis. Real-world priority should be calibrated by exposure surface: a public-facing comment system or documentation platform parsing untrusted Markdown is a reliable crash target, while an offline static site generator consuming author-controlled files carries negligible risk. The straightforward attack pattern and zero-prerequisite exploitation make this a meaningful operational risk for any service in the former category, despite the DoS-only impact ceiling. |
| Exploit ScenarioDerived | An attacker submits a POST request to a web application's comment or document-rendering endpoint with a Markdown body consisting of 1000 asterisks, a single letter, and 1000 more asterisks. Mistune 3.3.x builds an unbounded nested emphasis token tree from this input, and the recursive HTMLRenderer.render_token() call exhausts Python's default recursion limit, raising RecursionError and crashing the worker process or thread. The attack is repeatable and requires only HTTP access to any endpoint that renders user-supplied Markdown. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Aug 24, 2026
Denial-of-service in Mistune Python Markdown parser versions 3.3.0-3.3.2 allows remote unauthenticated attackers to crash any process that renders user-supplied Markdown by submitting input with hundreds of consecutive asterisk characters. The recursive HTMLRenderer.render_token() traversal grows proportionally to emphasis nesting depth and exceeds Python's default recursion limit, raising an unhandled RecursionError that terminates the parsing process. No public exploit identified at time of analysis, though the fix commit itself demonstrates the attack pattern explicitly in its regression test suite.
RemediationAI
The primary remediation is upgrading Mistune to version 3.3.3, which caps emphasis token nesting at a configurable depth (default 20) before recursive rendering can exhaust the Python call stack. The fix commit is at https://github.com/lepture/mistune/commit/0938fb7 and the security advisory at https://github.com/lepture/mistune/security/advisories/GHSA-6m44-fpc8-c3rq. If immediate upgrade is not feasible, two compensating controls apply with trade-offs: first, wrapping Markdown parsing calls in a try/except RecursionError block and returning an error response prevents process termination but still allows the DoS to consume a request slot and may mask unrelated recursion bugs elsewhere. Second, pre-filtering input to reject strings where any run of consecutive asterisk characters exceeds a threshold (e.g., 50) blocks the attack vector with low false-positive risk for legitimate Markdown, but requires careful implementation to avoid bypasses via mixed asterisk/underscore sequences. Upgrade to 3.3.3 is strongly preferred over either workaround.
Affected VersionsENISA
Affected Packages
Source: OSV (Open Source Vulnerabilities)
Affects your stack
Affects 1 package in pypi
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
Mistune's inline parser tokenizes Markdown emphasis constructs (*, , *) into a nested tree of 'emphasis' and 'strong' token objects. Prior to 3.3.3, _finalize_emphasis_tokens() and _process_emphasis_delimiters() imposed no upper bound on nesting depth, so input such as 1000 asterisks surrounding a single character produces a deeply recursive token tree. HTMLRenderer.render_token() traverses this tree with recursive calls; with sufficient nesting the Python call stack (default limit approximately 1000 frames) is exhausted, raising RecursionError. CWE-674 (Uncontrolled Recursion) is the precise root cause: recursion depth is directly proportional to attacker-controlled input without any defensive cap. The fix in 3.3.3 introduces a DEFAULT_MAX_EMPHASIS_DEPTH constant of 20, enforced in _process_emphasis_delimiters() via a new _emphasis_depth() iterative stack-walk that rejects emphasis pairs that would exceed the limit.
Know when Python gets its next vulnerability
- Watch products and get an email when something new is published
- Check this CVE against your own stack and servers
- Exploit and PoC references, threat intelligence from 22 sources
No credit card · 7-day full trial
Severity by source
AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
Network-delivered crafted input requires no auth or interaction; impact is strictly process availability loss with no confidentiality or integrity effect.
Primary rating from Vendor (github).
CVSS VectorVendor: github
Lifecycle Timeline
2DescriptionCVE.org
Mistune is a Python Markdown parser with renderers and plugins. Versions 3.3.0 through 3.3.2 are vulnerable to DoS through deeply nested tokens. HTML rendering creates deeply nested emphasis tokens from consecutive asterisk characters, and recursive rendering in HTMLRenderer.render_token() can exceed Python's recursion limit and raise RecursionError, allowing crafted Markdown to crash a parsing process. This issue is fixed in version 3.3.3
Blast Radius
ecosystem impact- 1 pypi packages depend on mistune (1 direct, 0 indirect)
Ecosystem-wide dependent count for version 3.3.0.
Articles & Coverage 1
Executive ViewAI
Mistune Python Markdown parser versions 3.3.0-3.3.2 contain a remote denial-of-service vulnerability allowing unauthenticated attackers to crash any application processing user-supplied Markdown content. Organizations using these versions face immediate availability risk, as attackers can terminate service processes without authentication by submitting specially formatted input.
No vendor-released patch identified at time of analysis. Implement input sanitization to detect and block Markdown with excessive consecutive punctuation characters; deploy WAF rules targeting payloads with hundreds of consecutive asterisks; enable rate-limiting on endpoints that process markdown; downgrade to Mistune 3.2.x or earlier if feasible; monitor application logs for RecursionError exceptions and unexpected process terminations.
Threat Intelligence Overview
How this decision was reached
The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.
What every action level means
SSVC AssessmentENISA
Priority Score
Attack Technique
Denial of Service attacks render applications or systems unavailable by overwhelming resources or triggering failure conditions.
Wazuh SIEM platform versions 4.4.0 through 4.9.0 contain an unsafe deserialization vulnerability in the DistributedAPI t
BentoML version 1.4.2 and earlier contains an unauthenticated remote code execution vulnerability through insecure deser
pgAdmin 4 contains critical remote code execution vulnerabilities in the Query Tool download and Cloud Deployment endpoi
The renderLocalView function in render/views.py in graphite-web in Graphite 0.9.5 through 0.9.10 uses the pickle Python
BentoML is a Python library for building online serving systems optimized for AI apps and model inference. Rated critica
OpenSSL before 0.9.8za, 1.0.0 before 1.0.0m, and 1.0.1 before 1.0.1h does not properly restrict processing of ChangeCiph
pyLoad download manager version prior to 0.5.0b3.dev77 exposes the Flask SECRET_KEY through an unauthenticated endpoint.
Langflow (a visual LLM pipeline builder) contains a critical unauthenticated code execution vulnerability (CVE-2026-3301
In Mercurial before 4.1.3, "hg serve --stdio" allows remote authenticated users to launch the Python debugger, and conse
Unauthenticated remote code execution affects Kestra OSS (the open-source event-driven orchestration platform) prior to
Unauthenticated remote code execution in Marimo ≤0.20.4 allows attackers to execute arbitrary system commands via the `/
pyLoad is the free and open-source Download Manager written in pure Python. Rated medium severity (CVSS 5.3), this vulne
EUVD-2026-65137
GHSA-6m44-fpc8-c3rq
NVD
CVE.org
EPSS
Vulnrichment
Exploit-DB