SQL injection in itsourcecode Courier Management System 1.0 via the ids parameter in /print_pdets.php allows authenticated remote attackers to execute arbitrary SQL queries with low impact to confidentiality, integrity, and availability. The vulnerability has a publicly available proof-of-concept and is characterized by low CVSS score (2.1) due to authentication requirement and limited scope, but SQL injection itself represents a persistent threat vector if the system processes sensitive courier data.
Missing authentication in Langchain-Chatchat up to version 0.3.1.3 allows unauthenticated local network attackers to access and manipulate files through the Compatible File Service endpoints (files/list_files, retrieve_file, retrieve_file_content, delete_file) in openai_routes.py without credentials. The vulnerability has a publicly available proof-of-concept exploit and affects confidentiality, integrity, and availability of stored files, though impact is limited to low severity per CVSS scoring; however, the lack of authentication on file operations represents a significant security control failure.
Insufficiently random file ID generation in Langchain-Chatchat up to version 0.3.1.3 allows authenticated local network attackers to predict uploaded file identifiers via the _get_file_id function, enabling information disclosure. The vulnerability requires local network access and authenticated privileges but carries low exploitability due to high attack complexity. A public exploit is available, though the project has not responded to early disclosure notifications.
Time-of-check time-of-use (TOCTOU) vulnerability in Langchain-Chatchat up to 0.3.1.3 allows authenticated local network attackers to manipulate file.filename arguments in the OpenAI-Compatible File Upload API, leading to integrity compromise through race condition exploitation. Publicly available exploit code exists, and the vendor has not responded to early disclosure notification despite proof-of-concept documentation.
Weak hash function in the Vision Chat Paste Image Handler of Langchain-Chatchat up to 0.3.1.3 allows local network attackers with low privileges to cause information disclosure via hash collision attacks on image data processed through PIL.Image.tobytes. The vulnerability requires high attack complexity and local network presence, resulting in minimal direct impact (CVSS 1.2); however, publicly available exploit code exists and the vendor has not responded to disclosure.
ciguard SCA HTTP clients in osv.py and endoflife.py exhaust process memory by reading unbounded response bodies from OSV.dev and endoflife.date without enforcing a size cap. A hostile or compromised backend service, or a successful TLS MITM attack, could return multi-gigabyte responses causing denial of service via out-of-memory kill or system swap exhaustion in CI environments. No public exploit identified at time of analysis; vendor released patch in v0.8.2.
HTTP response splitting in Microdot's Response.set_cookie() method allows header injection attacks when an attacker-controlled XSS payload reaches the server and is stored as a cookie value. The vulnerability stems from unsanitized carriage return and linefeed characters (\r\n) in cookie parameters, enabling an attacker to inject arbitrary HTTP headers. Exploitation requires prior client-side compromise (XSS), limiting the attack to a single compromised client per incident.
The discover_pipeline_files() function in ciguard v0.8.0-0.8.1 follows symlinks during directory traversal without validating that discovered paths remain within the scan root, allowing an attacker who plants symlinks in a scanned directory to cause the MCP server to leak file paths and contents from sensitive locations such as ~/.aws/, ~/.config/, and /etc/. This information disclosure vulnerability affects users of ciguard integrated with Claude Desktop, Claude Code, or Cursor, and is patched in v0.8.2 with default symlink following disabled and path validation applied.
The ciguard static analysis container image (versions 0.1.0-0.8.1) runs as root due to a missing USER directive in the Dockerfile, creating a privilege-escalation amplification risk for future container-runtime escape vulnerabilities. This is a defence-in-depth gap rather than a directly exploitable vulnerability; it reduces the impact of hypothetical escapes (such as runc CVE-2024-21626) from host-root compromise to non-root user compromise. Vendor-released patch in v0.8.2 adds a dedicated non-root ciguard user and USER directive, verified by container inspection and automated regression testing in v0.8.3.
Server-side request forgery (SSRF) in Geyser through version 2.9.2 allows authenticated attackers with operator privileges to cause the Minecraft server to issue arbitrary HTTP GET requests to internal or attacker-controlled endpoints via crafted Base64-encoded player head texture URLs in the /give command. The vulnerability enables blind SSRF attacks for network reconnaissance, cloud metadata probing, and server IP disclosure without requiring unauthenticated access. Publicly available exploit code exists demonstrating proof-of-concept via webhook.site.
Django 6.0 before 6.0.5 and 5.2 before 5.2.14 fail to vary response headers on session cookies when SESSION_SAVE_EVERY_REQUEST is enabled but the session is unmodified, allowing remote attackers with user interaction to steal session tokens from cached public pages. The vulnerability affects server configurations that cache responses aggressively while maintaining per-request session handling, exposing authenticated users to session hijacking after visiting pages served from cache.
OpenBao namespace deletion fails to properly clean up data and revoke leases when the initial deletion attempt is interrupted, potentially leaving orphaned storage entries and outstanding leases in the system. Subsequent retry attempts to delete the same namespace do not trigger proper cleanup, creating information disclosure and data integrity risks. This affects OpenBao versions prior to v2.5.3, with a vendor-released patch available.
Django's UpdateCacheMiddleware incorrectly caches HTTP responses containing a Vary header with an asterisk value in versions 6.0 before 6.0.5 and 5.2 before 5.2.14, causing private user data to be cached and served to other users. The vulnerability has low confidentiality impact and requires user interaction (UI:P) combined with passive attack timing, making real-world exploitation dependent on specific cache timing conditions and application architecture.
Race condition in Parse Server MFA SMS one-time password validation allows two concurrent login requests using the same OTP to both succeed and receive valid session tokens, breaking the single-use property of SMS-based multi-factor authentication. The vulnerability affects Parse Server versions 8.x before 8.6.76 and 9.x before 9.9.0-alpha.2, requires the attacker to already possess the victim's password and intercept the active SMS OTP via SIM swap, network interception, or phishing relay, then race a legitimate login request. This represents an incomplete fix of a prior optimistic locking vulnerability that affected only array-typed authData fields; SMS OTP storage as a string was not covered by the original guard.
Sandboxie-Plus versions 1.17.2 and earlier contain a cryptographic implementation flaw in SbieIniServer::HashPassword that reduces SHA-1 password hash entropy from 160 bits to 80 bits by incorrectly shifting the high nibble of each byte right by 8 instead of 4, combined with an unsalted hashing scheme. This makes leaked or backed-up EditPassword hashes significantly easier to brute-force, enabling attackers with local access and low privileges to recover plaintext passwords through offline attack. The vulnerability is fixed in version 1.17.3.
OpenClaw before 2026.4.10 allows local attackers with workspace write access to bypass boundary checks via a time-of-check-time-of-use race condition in the validateScriptFileForShellBleed function. An attacker can swap the target script file between validation and preflight read, causing the validator to inspect a different file than the one that passed the initial workspace boundary check, potentially leaking preflight metadata such as matched tokens or line numbers. No public exploit code identified at time of analysis; patch available in version 2026.4.10.