vLLM has Remote DoS via Invalid Recovered Token Reinjection

SummaryA frontend-legal multi-request speculative workload can make vLLM produce an out-of-vocabulary recovered token equal to vocabsize, convert that value to -1 when choosing the next live token for a request, and then feed that -1 back into the next drafter input ids. On Qwen3 GPTQ this reache...

LangBot: Authenticated RCE Via MCP Configuration

SummaryAny authenticated user can achieve arbitrary command execution on the LangBot servers through changing the MCP Server Configuration by added an "STDIO" MCP with an arbitrary command. DetailsThe repository uses StdioServerParameters which is based on Anthropic's modelcontextprotocol open...

OpenStack oslo.messaging does not verify RabbitMQ broker hostname during TLS handshake

An issue was discovered in OpenStack oslo.messaging 1.0.0 through 17.3.0. The oslo.messaging RabbitMQ driver does not perform TLS hostname verification when connecting to the message broker. When sslcafile is configured, the driver enables certificate chain validation but does not pass the expect...

PYSEC-2026-3455

pyasn1 is a generic ASN.1 library for Python. Prior to 0.6.4, the BER decoder shared by the CER and DER codecs parses long-form tags by accumulating continuation octets without an upper bound on the tag ID size, allowing a crafted input to force construction of an arbitrarily large integer with C...

Apache Airflow: Execution API JWT leaked via KubernetesExecutor worker command-line args

A bug in Apache Airflow's KubernetesExecutor caused JWT tokens used by worker pods to authenticate against the Execution API to be passed to the worker container as command-line arguments visible in the pod spec. An authenticated UI/API user with Kubernetes read-only access to the cluster e.g...

Shamefile has an arbitrary file read via shamefile.yaml in shame next

ImpactA path traversal vulnerability in shame next allows an attacker-controlled shamefile.yaml to disclose contents of files outside the repository, one line at a time, to the terminal of a user who runs the command. See patch commit for technical details. PatchesFixed in 0.1.7. Upgrade to eithe...

Apache Airflow FAB Auth Manager contains an LDAP filter injection vulnerability

Apache Airflow FAB Auth Manager contains an LDAP filter injection vulnerability CWE-90 that allows unauthenticated attackers to exfiltrate directory data or bypass authentication. Upgrade to apache-airflow-providers-fab 3.6.4 or later. If immediate upgrade is not possible, disable LDAP...

Open WebUI: Unauthenticated endpoint can trigger embedding generation (cost/DoS)

SummaryGET /api/v1/memories/ef is accessible without authentication and executes request.app.state.EMBEDDINGFUNCTION.... This allows any unauthenticated caller to trigger embedding generation which can lead to direct cost exposure if a paid provider is used.Code reference:...

PraisonAI knowledge-store backends interpolate unvalidated collection names into SQL and CQL queries

SummaryPraisonAI exposes optional SQL/CQL-backed knowledge-store implementations that build table and index identifiers from unvalidated name and collection arguments. Applications that pass untrusted collection names into these backends can trigger SQL or CQL injection. DetailsThis issue affects...

PraisonAI has unsafe tool resolution in `ToolExecutionMixin.execute_tool`: undeclared `__main__` callables execute

Summarypraisonaiagents resolves unresolved tool names against module globals and main after it fails to match the declared tool list and the registry. With the default agent configuration, permallow is None, so undeclared non-dangerous tool names are not rejected by the permission gate. An attack...

datapizza-ai: Server-Side Template Injection in ChatPromptTemplate via Jinja2 Template Handler

A flaw has been found in datapizza-labs datapizza-ai 0.0.2. Affected is the function ChatPromptTemplate of the file datapizza-ai-core/datapizza/modules/prompt/prompt.py of the component Jinja2 Template Handler. This manipulation of the argument Prompt causes improper neutralization of special...

Python Social Auth - Django has unsafe account association

ImpactUpon authentication, the user could be associated by e-mail even if the associatebyemail pipeline was not included. This could lead to account compromise when a third-party authentication service does not validate provided e-mail addresses or doesn't require unique e-mail addresses. Patches...

ExecuTorch integer overflow vulnerability leads to code execution

An integer overflow vulnerability in the loading of ExecuTorch models can cause smaller-than-expected memory regions to be allocated, potentially resulting in code execution or other undesirable effects. This issue affects ExecuTorch prior to commit 8f062d3f661e20bb19b24b767b9a9a46e8359f2b...

Apache HDFS Provider error message suggested

In the Apache Airflow HDFS Provider, versions prior to 4.1.1, a documentationinfo pointed users to an install incorrect pip package. As this package name was unclaimed, in theory, an attacker could claim this package and provide code that would be executed when this package was installed. The...

PYSEC-2026-179

PyJWT is a JSON Web Token implementation in Python. Prior to 2.13.0, when the verifier is decoding JSON Web Tokens, while supporting both asymmetric and HMAC algorithms, the library does not validate use of JSON Web Keys in HMAC algorithm, allowing attacker to use the issuer public key as the...

PYSEC-2026-180

Streamlink is a CLI utility which pipes video streams from various services into a video player. Prior to 8.4.0, Streamlink's HLS and DASH parsers do not validate the URI scheme of segment entries and other resources. A remote .m3u8 HLS playlist or .mpd DASH manifest can list file:///path/to/file...

PYSEC-0000-CVE-2026-44730

OpenCTI is an open source platform for managing cyber threat intelligence knowledge and observables. Prior to 6.9.7, an organization admin can escalate their privileges by adding a user from a different organization with higher privileges, to their own organization. This is due to incorrect ACL o...

PYSEC-2026-148

Wagtail is an open source content management system built on Django. Prior to 7.0.7, 7.3.2, and 7.4, a CMS user with limited access to form pages could delete submissions to form pages they don't have access to by crafting a form submission to delete submissions on a page they do have access to f...

PYSEC-2025-177

Pycel through 1.0b30, when operating on an untrusted spreadsheet, allows code execution via a crafted formula in a cell, such as one beginning with the =IFA1=200, eval"import'os'.system substring...

PYSEC-2020-44

This affects the package Gerapy from 0 and before 0.9.3. The input being passed to Popen, via the projectconfigure endpoint, isn’t being sanitized...

PYSEC-2014-59

Multiple open redirect vulnerabilities in 1 marmosetpatch.py, 2 publish.py, and 3 principiaredirect.py in Plone 2.1 through 4.1, 4.2.x through 4.2.5, and 4.3.x through 4.3.1 allow remote attackers to redirect users to arbitrary web sites and conduct phishing attacks via unspecified vectors...

AIOHTTP: HTTP request smuggling via WebSocket upgrade

SummaryThe HTTP parsers were vulnerable to a request smuggling attack relating to WebSocket upgrades. ImpactIf using the server-side component, it may be possible for an attacker to execute a request smuggling vulnerability using an edge case in the WebSocket upgrade procedure. AIOHTT is unaware ...

DIRAC is vulnerable to RCE in RequestManager due to eval on untrusted input

SummaryAn remote code execution vulnerability exists in RequestManager due to the use of eval on untrusted input that allows any authenticated user to run code/commands on the DIRAC server as the system user running the DIRAC services. DetailsThe exportgetRequestCountersWeb function is callable b...

Anki's local HTTP server does not sufficiently validate requests

SummaryAnki launches a local HTTP server to serve media files and web pages for parts of its interface. While the server has a CORS setup, requests from other origins were not blocked, allowing malicious websites to potentially trigger side-effecting requests. Browser impactThe severity varies by...

PraisonAI recipe serve Typer command bypasses the non-localhost authentication guard

PraisonAI recipe serve Typer command bypasses the non-localhost authentication guard SummaryPraisonAI's installed console entrypoint is Typer-first. In current releases,the recipe command is registered in the Typer app andpraisonai recipe serve dispatches to the deprecated Typer command...

PraisonAI: Compute-bridged file tools allow shell command injection

Compute-bridged file tools allow shell command injection SummaryLocalManagedAgent / SandboxedAgent compute bridging wrapsreadfile, listfiles, and writefile when a compute provider isattached. The bridge converts those file operations into shell command stringsusing raw path arguments, then sends...

OpenStack Ironic allows file overwrite via directory traversal during deployment with a crafted ISO image

OpenStack Ironic through before 35.0.2 allows file overwrite via directory traversal during deployment with a crafted ISO image...

`lxml_html_clean.Cleaner` does not strip `javascript:` URLs from namespaced URL attributes

lxmlhtmlclean.Cleaner does not strip javascript: URLs from namespaced URL attributes xlink:hrefReporter: Guillem Lefait · Date: 2026-05-10Affected: lxml ≤ 6.1.0 and lxmlhtmlclean ≤ 0.4.4 latest stableConfirmed against: lxml 6.1.0 + lxmlhtmlclean 0.4.4 on Python 3.13.5, 3.14.4, and 3.15.0a8 libxml...

aiograpi: Unsafe signup challenge path handling

aiograpi versions before 0.9.10 accepted server-supplied signup challenge paths and used them to build request URLs before validating that the paths were relative Instagram API paths. A malicious or tampered challenge payload could cause challenge handling requests to be sent outside the intended...

Open WebUI: Read-Only Users Can Toggle Note Pin Status via Incorrect Permission Check (Write via Read-Only Access)

SummaryThe POST /api/v1/notes/id/pin endpoint performs a write operation toggling the ispinned field but only checks for read permission. Users with read-only access to a shared note can pin/unpin it, which is a state-modifying action that should require write permission. All other write endpoint...

Open WebUI's Insecure Message Access Breaks Authorization

DescriptionThere's an IDOR in the channels message management system that allows authenticated users to modify or delete any message within channels they have read access to. The vulnerability exists in the message update and delete endpoints, which implement channel-level authorization but...

DiskCache has unsafe pickle deserialization

DiskCache python-diskcache through 5.6.3 uses Python pickle for serialization by default. An attacker with write access to the cache directory can achieve arbitrary code execution when a victim application reads from the cache...

Fickling has a bypass via runpy.run_path() and runpy.run_module()

Fickling's assessmentrunpy was added to the list of unsafe imports https://github.com/trailofbits/fickling/commit/9a2b3f89bd0598b528d62c10a64c1986fcb09f66. Original report SummaryFickling versions up to and including 0.1.6 do not treat Python’s runpy module as unsafe. Because of this, a malicious...

PYSEC-2026-181

A Dag author could either a create a symlink under their task's log directory pointing to an arbitrary file readable by the API server process read-path attack — e.g. /etc/passwd or airflow.cfg or b supply a taskid containing .. sequences accepted by the Task SDK's KEYREGEX write-path attack, and...

PYSEC-0000-CVE-2026-45017

Python Liquid is a Python engine for the Liquid template language. Prior to 2.2.0, the built-in FileSystemLoader and CachingFileSystemLoader do not guard against reading files outside their search paths when given an absolute path to resolve. This allows malicious template authors to load and...

PYSEC-2026-120

PyJWT is a JSON Web Token implementation in Python. Prior to 2.12.0, PyJWT does not validate the crit Critical Header Parameter defined in RFC 7515 §4.1.11. When a JWS token contains a crit array listing extensions that PyJWT does not understand, the library accepts the token instead of rejecting...

PYSEC-2026-43

An issue was discovered in 6.0 before 6.0.2, 5.2 before 5.2.11, and 4.2 before 4.2.28.ASGIRequest allows a remote attacker to cause a potential denial-of-service via a crafted request with multiple duplicate headers.Earlier, unsupported Django series such as 5.0.x, 4.1.x, and 3.2.x were not...

Open WebUI: Terminal proxy forwards a spoofable, integrity-unbound user identity to the upstream (X-User-Id header and ws_terminal session_id query injection)

SummaryThe terminal proxy in backend/openwebui/routers/terminals.py forwards the Open WebUI user's identity to the upstream terminal server / backend coordinator as an authorization claim, with no cryptographic binding to the session that produced it. The forwarded identity is...

Pillow JPEG2000 tiled decode retains a growing scratch buffer and can be used for denial of service

Summarysrc/libImaging/Jpeg2KDecode.c:853 accumulates totalcomponentwidth across every tile in a JPEG2000 image instead of recomputing it per tile. That accumulated value is then used in the tilebytes calculation at src/libImaging/Jpeg2KDecode.c:868, which can make the decoder grow state-buffer vi...

Prompty: Arbitrary file read via file reference expansion

SummaryPrompty loaders expanded $file:... references in .prompty frontmatter without enforcing that the resolved path stayed within an authorized directory. An attacker-controlled prompt file could use path traversal or an absolute path to cause the host application to read files accessible to th...

PraisonAI: HTTPApproval dashboard renders tool arguments as raw HTML, allowing approval-page XSS to approve dangerous tools

HTTPApproval dashboard renders tool arguments as raw HTML, allowing approval-page XSS to approve dangerous tools Summarypraisonai.bots.HTTPApproval renders pending tool approval arguments directlyinto the approval dashboard HTML. An attacker-controlled tool argument caninject JavaScript into that...

OpenStack Mistral allows Arbitrary Remote Code Execution when the API is exposed

OpenStack Mistral through 22.0.0 allows Arbitrary Remote Code Execution when the API is exposed. There are endpoints that allow code execution, which can lead to exfiltration of service credentials...

vllm has Improper Resource Shutdown or Release

A vulnerability was identified in vllm-project vllm 0.19.0. This issue affects some unknown processing of the component OpenAI-compatible Serving Path. Such manipulation leads to denial of service. It is possible to launch the attack remotely. The exploit is publicly available and might be used...

curl_cffi: Redirect-based SSRF leads to internal network access in curl_cffi (with TLS impersonation bypass)

Summarycurlcffi does not restrict requests to internal IP ranges, and follows redirects automatically via the underlying libcurl.Because of this, an attacker-controlled URL can redirect requests to internal services such as cloud metadata endpoints. In addition, curlcffi’s TLS impersonation featu...

Salt Authentication Protocol Version Downgrade Allows Minion Impersonation

Salt contains an authentication protocol version downgrade weakness that can allow a malicious minion to bypass newer authentication/security features by using an older request payload format, enabling minion impersonation and circumventing protections introduced in response to prior issues...

Keras Directory Traversal Vulnerability

SummaryKeras's keras.utils.getfile function is vulnerable to directory traversal attacks despite implementing filtersafepaths. The vulnerability exists because extractarchive uses Python's tarfile.extractall method without the security-critical filter="data" parameter. A PATHMAX symlink resolutio...

Inefficient Regular Expression Complexity in langflow

A vulnerability classified as problematic was found in Langflow up to 1.0.18. Affected by this vulnerability is an unknown functionality of the file \src\backend\base\langflow\interface\utils.py of the component HTTP POST Request Handler. The manipulation of the argument remainingtext leads to...

SQL injection in litellm

An SQL Injection vulnerability exists in the berriai/litellm repository, specifically within the /global/spend/logs endpoint. The vulnerability arises due to improper neutralization of special elements used in an SQL command. The affected code constructs an SQL query by concatenating an unvalidat...

ansibleguy-webui Cross-site Scripting vulnerability

ImpactMultiple forms in version = 0.0.21 References Report GitHub Issue 44...

Open Redirect in django-spirit

django-spirit prior to version 0.12.3 is vulnerable to open redirect. In the /user/login endpoint, it doesn't check the value of the next parameter when the user is logged in and passes it directly to redirect which result to open redirect. This also affects /user/logout, /user/register,...