OpenClaw versions prior to 2026.3.1 contain an unbounded memory growth vulnerability in the Zalo webhook endpoint that allows unauthenticated attackers to trigger in-memory key accumulation by varying query strings. Remote attackers can exploit this by sending repeated requests with different query parameters to cause memory pressure, process instability, or out-of-memory conditions that degrade service availability.
OpenClaw versions prior to 2026.2.22 contain an allowlist bypass vulnerability in system.run that allows attackers to execute non-allowlisted commands by splitting command substitution using shell line-continuation characters. Attackers can bypass security analysis by injecting $\\ followed by a newline and opening parenthesis inside double quotes, causing the shell to fold the line continuation into executable command substitution that circumvents approval boundaries.
OpenClaw versions prior to 2026.2.25 lack durable replay state for Nextcloud Talk webhook events, allowing valid signed webhook requests to be replayed without suppression. Attackers can capture and replay previously valid signed webhook requests to trigger duplicate inbound message processing and cause integrity or availability issues.
OpenClaw versions prior to 2026.3.2 contain a race condition vulnerability in ZIP extraction that allows local attackers to write files outside the intended destination directory. Attackers can exploit a time-of-check-time-of-use race between path validation and file write operations by rebinding parent directory symlinks to redirect writes outside the extraction root.
OpenClaw versions prior to 2026.2.22 contain an allowlist bypass vulnerability in system.run exec analysis that fails to unwrap env and shell-dispatch wrapper chains. Attackers can route execution through wrapper binaries like env bash to smuggle payloads that satisfy allowlist entries while executing non-allowlisted commands.
OpenClaw versions prior to 2026.2.19 contain a command injection vulnerability in Windows Scheduled Task script generation where environment variables are written to gateway.cmd using unquoted set KEY=VALUE assignments, allowing shell metacharacters to break out of assignment context. Attackers can inject arbitrary commands through environment variable values containing metacharacters like &, |, ^, %, or ! to achieve command execution when the scheduled task script is generated and executed.
PX4 is an open-source autopilot stack for drones and unmanned vehicles. Versions 1.17.0-rc2 and below are vulnerable to Stack-based Buffer Overflow through the MavlinkLogHandler, and are triggered via MAVLink log request. The LogEntry.filepath buffer is 60 bytes, but the sscanf function parses paths from the log list file with no width specifier, allowing a path longer than 60 characters to overflow the buffer. An attacker with MAVLink link access can trigger this by first creating deeply nested directories via MAVLink FTP, then requesting the log list. The flight controller MAVLink task crashes, losing telemetry and command capability and causing DoS. This issue has been fixed in this commit: https://github.com/PX4/PX4-Autopilot/commit/616b25a280e229c24d5cf12a03dbf248df89c474.
Kan is an open-source project management tool. In versions 0.5.4 and below, the /api/download/attatchment endpoint has no authentication and no URL validation. The Attachment Download endpoint accepts a user-supplied URL query parameter and passes it directly to fetch() server-side, and returns the full response body. An unauthenticated attacker can use this to make HTTP requests from the server to internal services, cloud metadata endpoints, or private network resources. This issue has been fixed in version 0.5.5. To workaround this issue, block or restrict access to /api/download/attatchment at the reverse proxy level (nginx, Cloudflare, etc.).
Romeo gives the capability to reach high code coverage of Go ≥1.20 apps by helping to measure code coverage for functional and integration tests within GitHub Actions. Prior to version 0.2.2, the `sanitizeArchivePath` function in `webserver/api/v1/decoder.go` (lines 80-88) is vulnerable to a path traversal bypass due to a missing trailing path separator in the `strings.HasPrefix` check. A crafted tar archive can write files outside the intended destination directory. Version 0.2.2 fixes the issue.
Romeo gives the capability to reach high code coverage of Go ≥1.20 apps by helping to measure code coverage for functional and integration tests within GitHub Actions. Prior to version 0.2.1, due to a mis-written NetworkPolicy, a malicious actor can pivot from the "hardened" namespace to any Pod out of it. This breaks the security-by-default property expected as part of the deployment program, leading to a potential lateral movement. Removing the `inter-ns` NetworkPolicy patches the vulnerability in version 0.2.1. If updates are not possible in production environments, manually delete `inter-ns` and update as soon as possible. Given one's context, delete the failing network policy that should be prefixed by `inter-ns-` in the target namespace.
The Hytale Modding Wiki is a free service for Hytale mods to host their documentation & wikis. An Insecure Direct Object Reference (IDOR) vulnerability in versions of the wiki prior to 1.0.0 exposes mod authors' personal information - including full names and email addresses - to any authenticated user who visits a mod page. Any user who creates an account can access sensitive author details by simply navigating to a mod's page via its slug. Version 1.0.0 fixes the issue.
openapi-to-java-records-mustache-templates allows users to generate Java Records from OpenAPI specifications. Starting in version 5.1.1 and prior to version 5.5.1, the parent POM file of this project (`openapi-to-java-records-mustache-templates-parent`), which is used to centralize plugin configurations for multiple unit-test modules, uses `maven-dependency-plugin` to unpack arbitrary `.mustache` files from the `openapi-to-java-records-mustache-templates` artifact (of the same version). While this parent POM file is not intended for external use, it is published, and could be used by anyone, and does not follow the best security practices. The risk, is that if `openapi-to-java-records-mustache-templates` would be compromised, and malicious `.mustache` files were to be included in the resulting JAR/artifact, users would unpack these files automatically during a dependency update. This is addressed in the v3.5.1 release of `openapi-to-java-records-mustache-templates-parent`. It is strongly recommended NOT to use the parent POM for external use. The `openapi-to-java-records-mustache-templates` module is the center of this project, and surrounding modules and configurations are not intended for production-use. These only exist for testing purposes and maintainability.
ApostropheCMS is an open-source content management framework. Prior to version 3.5.3 of `@apostrophecms/import-export`,
The `extract()` function in `gzip.js` constructs file-write paths using `fs.createWriteStream(path.join(exportPath, header.name))`. `path.join()` does not resolve or sanitise traversal segments such as `../`. It concatenates them as-is, meaning a tar entry named `../../evil.js` resolves to a path outside the intended extraction directory. No canonical-path check is performed before the write stream is opened. This is a textbook Zip Slip vulnerability. Any user who has been granted the Global Content Modify permission — a role routinely assigned to content editors and site managers — can upload a crafted `.tar.gz` file through the standard CMS import UI and write attacker-controlled content to any path the Node.js process can reach on the host filesystem. Version 3.5.3 of `@apostrophecms/import-export` fixes the issue.
ApostropheCMS is an open-source content management framework. Prior to version 4.28.0, the bearer token authentication middleware in `@apostrophecms/express/index.js` (lines 386-389) contains an incorrect MongoDB query that allows incomplete login tokens — where the password was verified but TOTP/MFA requirements were NOT — to be used as fully authenticated bearer tokens. This completely bypasses multi-factor authentication for any ApostropheCMS deployment using `@apostrophecms/login-totp` or any custom `afterPasswordVerified` login requirement. Version 4.28.0 fixes the issue.
A vulnerability in MLflow's pyfunc extraction process allows for arbitrary file writes due to improper handling of tar archive entries. Specifically, the use of `tarfile.extractall` without path validation enables crafted tar.gz files containing `..` or absolute paths to escape the intended extraction directory. This issue affects the latest version of MLflow and poses a high/critical risk in scenarios involving multi-tenant environments or ingestion of untrusted artifacts, as it can lead to arbitrary file overwrites and potential remote code execution.
Parse Server is an open source backend that can be deployed to any infrastructure that can run Node.js. Prior to 9.6.0-alpha.35 and 8.6.50, when a `Parse.Cloud.afterLiveQueryEvent` trigger is registered for a class, the LiveQuery server leaks protected fields and `authData` to all subscribers of that class. Fields configured as protected via Class-Level Permissions (`protectedFields`) are included in LiveQuery event payloads for all event types (create, update, delete, enter, leave). Any user with sufficient CLP permissions to subscribe to the affected class can receive protected field data of other users, including sensitive personal information and OAuth tokens from third-party authentication providers. The vulnerability was caused by a reference detachment bug. When an `afterEvent` trigger is registered, the LiveQuery server converts the event object to a `Parse.Object` for the trigger, then creates a new JSON copy via `toJSONwithObjects()`. The sensitive data filter was applied to the `Parse.Object` reference, but the unfiltered JSON copy was sent to clients. The fix in versions 9.6.0-alpha.35 and 8.6.50 ensures that the JSON copy is assigned back to the response object before filtering, so the filter operates on the actual data sent to clients. As a workaround, remove all `Parse.Cloud.afterLiveQueryEvent` trigger registrations. Without an `afterEvent` trigger, the reference detachment does not occur and protected fields are correctly filtered.
Parse Server is an open source backend that can be deployed to any infrastructure that can run Node.js. Prior to 9.6.0-alpha.29 and 8.6.49, a user can sign up without providing credentials by sending an empty `authData` object, bypassing the username and password requirement. This allows the creation of authenticated sessions without proper credentials, even when anonymous users are disabled. The fix in 9.6.0-alpha.29 and 8.6.49 ensures that empty or non-actionable `authData` is treated the same as absent `authData` for the purpose of credential validation on new user creation. Username and password are now required when no valid auth provider data is present. As a workaround, use a Cloud Code `beforeSave` trigger on the `_User` class to reject signups where `authData` is empty and no username/password is provided.
Parse Server is an open source backend that can be deployed to any infrastructure that can run Node.js. Prior to 9.6.0-alpha.21 and 8.6.45, an unauthenticated attacker can crash the Parse Server process by sending a single request with deeply nested query condition operators. This terminates the server and denies service to all connected clients. Starting in version 9.6.0-alpha.21 and 8.6.45, a depth limit for query condition operator nesting has been added via the `requestComplexity.queryDepth` server option. The option is disabled by default to avoid a breaking change. To mitigate, upgrade and set the option to a value appropriate for your app. No known workarounds are available.
Parse Server is an open source backend that can be deployed to any infrastructure that can run Node.js. Prior to 9.6.0-alpha.28 and 8.6.48, the password reset mechanism does not enforce single-use guarantees for reset tokens. When a user requests a password reset, the generated token can be consumed by multiple concurrent requests within a short time window. An attacker who has intercepted a password reset token can race the legitimate user's password reset request, causing both requests to succeed. This may result in the legitimate user believing their password was changed successfully while the attacker's password takes effect instead. All Parse Server deployments that use the password reset feature are affected. Starting in versions 9.6.0-alpha.28 and 8.6.48, the password reset token is now atomically validated and consumed as part of the password update operation. The database query that updates the password includes the reset token as a condition, ensuring that only one concurrent request can successfully consume the token. Subsequent requests using the same token will fail because the token has already been cleared. There is no known workaround other than upgrading.
Parse Server is an open source backend that can be deployed to any infrastructure that can run Node.js. Prior to 9.6.0-alpha.24 and 8.6.47, remote clients can crash the Parse Server process by calling a cloud function endpoint with a crafted function name that traverses the JavaScript prototype chain of a registered cloud function handler, causing a stack overflow. The fix in versions 9.6.0-alpha.24 and 8.6.47 restricts property lookups during cloud function name resolution to own properties only, preventing prototype chain traversal from stored function handlers. There is no known workaround.
Parse Server is an open source backend that can be deployed to any infrastructure that can run Node.js. Prior to 9.6.0-alpha.20 and 8.6.44, an attacker can bypass the default request keyword denylist protection and the class-level permission for adding fields by sending a crafted request that exploits prototype pollution in the deep copy mechanism. This allows injecting fields into class schemas that have field addition locked down, and can cause permanent schema type conflicts that cannot be resolved even with the master key. In 9.6.0-alpha.20 and 8.6.44, the vulnerable third-party deep copy library has been replaced with a built-in deep clone mechanism that handles prototype properties safely, allowing the existing denylist check to correctly detect and reject the prohibited keyword. No known workarounds are available.
Parse Server is an open source backend that can be deployed to any infrastructure that can run Node.js. Prior to 9.6.0-alpha.19 and 8.6.43, a remote attacker can crash the Parse Server by subscribing to a LiveQuery with an invalid regular expression pattern. The server process terminates when the invalid pattern reaches the regex engine during subscription matching, causing denial of service for all connected clients. The fix in 9.6.0-alpha.19 and 8.6.43 validates regular expression patterns at subscription time, rejecting invalid patterns before they are stored. Additionally, a defense-in-depth try-catch prevents any subscription matching error from crashing the server process. As a workaround, disable LiveQuery if it is not needed.
Parse Server is an open source backend that can be deployed to any infrastructure that can run Node.js. Prior to 9.6.0-alpha.17 and 8.6.42, an authenticated user can overwrite server-generated session fields (`sessionToken`, `expiresAt`, `createdWith`) when creating a session object via `POST /classes/_Session`. This allows bypassing the server's session expiration policy by setting an arbitrary far-future expiration date. It also allows setting a predictable session token value. Starting in version 9.6.0-alpha.17 and 8.6.42, the session creation endpoint filters out server-generated fields from user-supplied data, preventing them from being overwritten. As a workaround, add a `beforeSave` trigger on the `_Session` class to validate and reject or strip any user-supplied values for `sessionToken`, `expiresAt`, and `createdWith`.
Parse Server is an open source backend that can be deployed to any infrastructure that can run Node.js. Prior to 9.6.0-alpha.15 and 8.6.41, an attacker who is allowed to upload files can bypass the file extension filter by appending a MIME parameter (e.g. `;charset=utf-8`) to the `Content-Type` header. This causes the extension validation to fail matching against the blocklist, allowing active content to be stored and served under the application's domain. In addition, certain XML-based file extensions that can render scripts in web browsers are not included in the default blocklist. This can lead to stored XSS attacks, compromising session tokens, user credentials, or other sensitive data accessible via the browser's local storage. The fix in versions 9.6.0-alpha.15 and 8.6.41 strips MIME parameters from the `Content-Type` header before validating the file extension against the blocklist. The default blocklist has also been extended to include additional XML-based extensions (`xsd`, `rng`, `rdf`, `rdf+xml`, `owl`, `mathml`, `mathml+xml`) that can render active content in web browsers. Note that the `fileUpload.fileExtensions` option is intended to be configured as an allowlist of file extensions that are valid for a specific application, not as a denylist. The default denylist is provided only as a basic default that covers most common problematic extensions. It is not intended to be an exhaustive list of all potentially dangerous extensions. Developers should not rely on the default value, as new extensions that can render active content in browsers might emerge in the future. As a workaround, configure the `fileUpload.fileExtensions` option to use an allowlist of only the file extensions that your application needs, rather than relying on the default blocklist.
SandboxJS is a JavaScript sandboxing library. Prior to 0.8.35, SandboxJS timers have an execution-quota bypass. A global tick state (`currentTicks.current`) is shared between sandboxes. Timer string handlers are compiled at execution time using that global tick state rather than the scheduling sandbox's tick object. In multi-tenant / concurrent sandbox scenarios, another sandbox can overwrite `currentTicks.current` between scheduling and execution, causing the timer callback to run under a different sandbox's tick budget and bypass the original sandbox's execution quota/watchdog. Version 0.8.35 fixes this issue.
Memray is a memory profiler for Python. Prior to Memray 1.19.2, Memray rendered the command line of the tracked process directly into generated HTML reports without escaping. Because there was no escaping, attacker-controlled command line arguments were inserted as raw HTML into the generated report. This allowed JavaScript execution when a victim opened the generated report in a browser. Version 1.19.2 fixes the issue.
OpenProject is an open-source, web-based project management software. In versions prior to 16.6.9, 17.0.6, 17.1.3, and 17.2.1, the Repositories module did not properly escape filenames displayed from repositories. This allowed an attacker with push access into the repository to create commits with filenames that included HTML code that was injected in the page without proper sanitation. This allowed a persisted XSS attack against all members of this project that accessed the repositories page to display a changeset where the maliciously crafted file was deleted. Versions 16.6.9, 17.0.6, 17.1.3, and 17.2.1 fix the issue.
OpenProject is an open-source, web-based project management software. Versions prior to 16.6.9, 17.0.6, 17.1.3, and 17.2.1 are vulnerable to an SQL injection attack via a custom field's name. When that custom field was used in a Cost Report, the custom field's name was injected into the SQL query without proper sanitation. This allowed an attacker to execute arbitrary SQL commands during the generation of a Cost Report. As custom fields can only be generated by users with full administrator privileges, the attack surface is somewhat reduced. Together with another bug in the Repositories_module, that used the project identifier without sanitation to generate the checkout path for a git repository in the filesystem, this allowed an attacker to checkout a git repository to an arbitrarily chosen path on the server. If the checkout is done within certain paths within the OpenProject application, upon the next restart of the application, this allows the attacker to inject ruby code into the application. As the project identifier cannot be manually edited to any string containing special characters like dots or slashes, this needs to be changed via the SQL injection described above. Versions 16.6.9, 17.0.6, 17.1.3, and 17.2.1 fix the issue.
Devise is an authentication solution for Rails based on Warden. Prior to version 5.0.3, a race condition in Devise's Confirmable module allows an attacker to confirm an email address they do not own. This affects any Devise application using the `reconfirmable` option (the default when using Confirmable with email changes). By sending two concurrent email change requests, an attacker can desynchronize the `confirmation_token` and `unconfirmed_email` fields. The confirmation token is sent to an email the attacker controls, but the `unconfirmed_email` in the database points to a victim's email address. When the attacker uses the token, the victim's email is confirmed on the attacker's account. This is patched in Devise v5.0.3. Users should upgrade as soon as possible. As a workaround, applications can override a specific method from Devise models to force `unconfirmed_email` to be persisted when unchanged. Note that Mongoid does not seem to respect that `will_change!` should force the attribute to be persisted, even if it did not really change, so the user might have to implement a workaround similar to Devise by setting `changed_attributes["unconfirmed_email"] = nil` as well.
StudioCMS is a server-side-rendered, Astro native, headless content management system. Prior to 0.4.4, the REST API `getUsers` endpoint in StudioCMS uses the attacker-controlled `rank` query parameter to decide whether owner accounts should be filtered from the result set. As a result, an admin token can request `rank=owner` and receive owner account records, including IDs, usernames, display names, and email addresses, even though the adjacent `getUser` endpoint correctly blocks admins from viewing owner users. This is an authorization inconsistency inside the same user-management surface. Version 0.4.4 fixes the issue.
ImageMagick is free and open-source software used for editing and manipulating digital images. Prior to 7.1.2-17 and 6.9.13-42, the NewXMLTree method contains a bug that could result in a crash due to an out of write bounds of a single zero byte. Versions 7.1.2-17 and 6.9.13-42 fix the issue.
ClipBucket v5 is an open source video sharing platform. An authenticated time-based blind SQL injection vulnerability exists in ClipBucket prior to 5.5.3 #80 within the `actions/ajax.php` endpoint. Due to insufficient input sanitization of the `userid` parameter, an authenticated attacker can execute arbitrary SQL queries, leading to full database disclosure and potential administrative account takeover. Version 5.5.3 #80 fixes the issue.
SAMtools is a program for reading, manipulating and writing bioinformatics file formats. Starting in version 1.17, in the cram-size command, used to write information about how well CRAM files are compressed, a check to see if the `cram_decode_compression_header()` was missing. If the function returned an error, this could lead to a NULL pointer dereference. Exploiting this bug causes a NULL pointer dereference. Typically this will cause the program to crash. Versions 1.23.1, 1.22.2 and 1.21.1 include fixes for this issue. There is no workaround for this issue.
SAMtools is a program for reading, manipulating and writing bioinformatics file formats. The `mpileup` command outputs DNA sequences that have been aligned against a known reference. On each output line it writes the reference position, optionally the reference DNA base at that position (obtained from a separate file) and all of the DNA bases that aligned to that position. As the output is ordered by position, reference data that is no longer needed is discarded once it has been printed out. Under certain conditions the data could be discarded too early, leading to an attempt to read from a pointer to freed memory. This bug may allow information about program state to be leaked. It may also cause a program crash through an attempt to access invalid memory. This bug is fixed in versions 1.21.1 and 1.22. There is no workaround for this issue.
OmniGen2-RL contains an unauthenticated remote code execution vulnerability in the reward server component that allows remote attackers to execute arbitrary commands by sending malicious HTTP POST requests. Attackers can exploit insecure pickle deserialization of request bodies to achieve code execution on the host system running the exposed service.
OpenEMR is a free and open source electronic health records and medical practice management application. In versions up to and including 8.0.0, the message/note update endpoint (e.g. PUT or POST) updates by message/note ID only and does not verify that the message belongs to the current patient (or that the user is allowed to edit that patient’s notes). An authenticated user with notes permission can modify any patient’s messages by supplying another message ID. Commit 92a2ff9eaaa80674b3a934a6556e35e7aded5a41 contains a fix for the issue.
Improper certificate validation in Devolutions Hub Reporting Service
2025.3.1.1 and earlier allows a network attacker to perform a
man-in-the-middle attack via disabled TLS certificate verification.
HTSlib is a library for reading and writing bioinformatics file formats. CRAM is a compressed format which stores DNA sequence alignment data using a variety of encodings and compression methods. When reading data encoded using the `BYTE_ARRAY_LEN` method, the `cram_byte_array_len_decode()` failed to validate that the amount of data being unpacked matched the size of the output buffer where it was to be stored. Depending on the data series being read, this could result either in a heap or a stack overflow with attacker-controlled bytes. Depending on the data stream this could result either in a heap buffer overflow or a stack overflow. If a user opens a file crafted to exploit this issue it could lead to the program crashing, overwriting of data structures on the heap or stack in ways not expected by the program, or changing the control flow of the program. It may be possible to use this to obtain arbitrary code execution. Versions 1.23.1, 1.22.2 and 1.21.1 include fixes for this issue. There is no workaround for this issue.
HTSlib is a library for reading and writing bioinformatics file formats. GZI files are used to index block-compressed GZIP [BGZF] files. In the GZI loading function, `bgzf_index_load_hfile()`, it was possible to trigger an integer overflow, leading to an under- or zero-sized buffer being allocated to store the index. Sixteen zero bytes would then be written to this buffer, and, depending on the result of the overflow the rest of the file may also be loaded into the buffer as well. If the function did attempt to load the data, it would eventually fail due to not reading the expected number of records, and then try to free the overflowed heap buffer. Exploiting this bug causes a heap buffer overflow. If a user opens a file crafted to exploit this issue, it could lead to the program crashing, or overwriting of data and heap structures in ways not expected by the program. It may be possible to use this to obtain arbitrary code execution. Versions 1.23.1, 1.22.2 and 1.21.1 include fixes for this issue. The easiest work-around is to discard any `.gzi` index files from untrusted sources, and use the `bgzip -r` option to recreate them.
HTSlib is a library for reading and writing bioinformatics file formats. CRAM is a compressed format which stores DNA sequence alignment data using a variety of encodings and compression methods. When reading data encoded using the `BYTE_ARRAY_STOP` method, an out-by-one error in the `cram_byte_array_stop_decode_char()` function check for a full output buffer could result in a single attacker-controlled byte being written beyond the end of a heap allocation. Exploiting this bug causes a heap buffer overflow. If a user opens a file crafted to exploit this issue, it could lead to the program crashing, or overwriting of data and heap structures in ways not expected by the program. It may be possible to use this to obtain arbitrary code execution. Versions 1.23.1, 1.22.2 and 1.21.1 include fixes for this issue. There is no workaround for this issue.
HTSlib is a library for reading and writing bioinformatics file formats. CRAM is a compressed format which stores DNA sequence alignment data using a variety of encodings and compression methods. For the `VARINT` and `CONST` encodings, incomplete validation of the context in which the encodings were used could result in up to eight bytes being written beyond the end of a heap allocation, or up to eight bytes being written to the location of a one byte variable on the stack, possibly causing the values to adjacent variables to change unexpectedly. Depending on the data stream this could result either in a heap buffer overflow or a stack overflow. If a user opens a file crafted to exploit this issue it could lead to the program crashing, overwriting of data structures on the heap or stack in ways not expected by the program, or changing the control flow of the program. It may be possible to use this to obtain arbitrary code execution. Versions 1.23.1, 1.22.2 and 1.21.1 include fixes for this issue. There is no workaround for this issue.
HTSlib is a library for reading and writing bioinformatics file formats. CRAM is a compressed format which stores DNA sequence alignment data. In the `cram_decode_slice()` function called while reading CRAM records, the value of the mate reference id field was not validated. Later use of this value, for example when converting the data to SAM format, could result in the out of bounds array reads when looking up the corresponding reference name. If the array value obtained also happened to be a valid pointer, it would be interpreted as a string and an attempt would be made to write the data as part of the SAM record. This bug may allow information about program state to be leaked. It may also cause a program crash through an attempt to access invalid memory. Versions 1.23.1, 1.22.2 and 1.21.1 include fixes for this issue. There is no workaround for this issue.
HTSlib is a library for reading and writing bioinformatics file formats. CRAM is a compressed format which stores DNA sequence alignment data. As one method of removing redundant data, CRAM uses reference-based compression so that instead of storing the full sequence for each alignment record it stores a location in an external reference sequence along with a list of differences to the reference at that location as a sequence of "features". When decoding CRAM records, the reference data is stored in a char array, and parts matching the alignment record sequence are copied over as necessary. Due to insufficient validation of the feature data series, it was possible to make the `cram_decode_seq()` function copy data from either before the start, or after the end of the stored reference either into the buffer used to store the output sequence for the cram record, or into the buffer used to build the SAM `MD` tag. This allowed arbitrary data to be leaked to the calling function. This bug may allow information about program state to be leaked. It may also cause a program crash through an attempt to access invalid memory. Versions 1.23.1, 1.22.2 and 1.21.1 include fixes for this issue. There is no workaround for this issue.
Rejected reason: After the publication of the PoC by the researcher and further analysis, we have determined that this issue does not constitute a valid vulnerability. The technique described is an obfuscation method and does not bypass or impact any implicit or explicit security controls.
HTSlib is a library for reading and writing bioinformatics file formats. CRAM is a compressed format which stores DNA sequence alignment data. In the `cram_decode_slice()` function called while reading CRAM records, validation of the reference id field occurred too late, allowing two out of bounds reads to occur before the invalid data was detected. The bug does allow two values to be leaked to the caller, however as the function reports an error it may be difficult to exploit them. It is also possible that the program will crash due to trying to access invalid memory. Versions 1.23.1, 1.22.2 and 1.21.1 include fixes for this issue. There is no workaround for this issue.
HTSlib is a library for reading and writing bioinformatics file formats. CRAM is a compressed format which stores DNA sequence alignment data using a variety of encodings and compression methods. While most alignment records store DNA sequence and quality values, the format also allows them to omit this data in certain cases to save space. Due to some quirks of the CRAM format, it is necessary to handle these records carefully as they will actually store data that needs to be consumed and then discarded. Unfortunately the `CONST`, `XPACK` and `XRLE` encodings did not properly implement the interface needed to do this. Trying to decode records with omitted sequence or quality data using these encodings would result in an attempt to write to a NULL pointer. Exploiting this bug causes a NULL pointer dereference. Typically this will cause the program to crash. Versions 1.23.1, 1.22.2 and 1.21.1 include fixes for this issue. There is no workaround for this issue.
HTSlib is a library for reading and writing bioinformatics file formats. CRAM is a compressed format which stores DNA sequence alignment data. As one method of removing redundant data, CRAM uses reference-based compression so that instead of storing the full sequence for each alignment record it stores a location in an external reference sequence along with a list of differences to the reference at that location as a sequence of "features". When decoding these features, an out-by-one error in a test for CRAM features that appear beyond the extent of the CRAM record sequence could result in an invalid write of one attacker-controlled byte beyond the end of a heap buffer. Exploiting this bug causes a heap buffer overflow. If a user opens a file crafted to exploit this issue, it could lead to the program crashing, or overwriting of data and heap structures in ways not expected by the program. It may be possible to use this to obtain arbitrary code execution. Versions 1.23.1, 1.22.2 and 1.21.1 include fixes for this issue. There is no workaround for this issue.