EUVD-2026-25855Severity by source
AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H
Unauthenticated remote POST to a reachable search endpoint (AV:N/AC:L/PR:N/UI:N) causes resource exhaustion affecting only availability, so C:N/I:N/A:H with unchanged scope.
Primary rating from NVD.
CVSS VectorNVD
Lifecycle Timeline
4DescriptionCVE.org
An issue in the /store/items/search endpoint of Agent Protocol server commit e9a89f allows attackers to cause a Denial of Service (DoS) via a crafted POST request.
AnalysisAI
Denial of service in Aegra, an open-source Agent Protocol server (vulnerable at commit e9a89f), lets remote unauthenticated attackers exhaust server resources by sending a crafted POST request to the /store/items/search endpoint. The flaw is classified as uncontrolled resource consumption (CWE-400) and carries CVSS 7.5 with an availability-only impact; publicly available exploit code exists via a referenced gist, but there is no evidence of active exploitation and the EPSS probability is negligible (0.01%).
Technical ContextAI
Aegra is a self-hosted implementation of the Agent Protocol, an HTTP API surface used to drive AI agent workflows (runs, threads, and a persistence 'store' for items). The affected /store/items/search endpoint accepts POST bodies used to query stored items; the underlying weakness is CWE-400 (Uncontrolled Resource Consumption), meaning the endpoint fails to bound the work, memory, or query cost triggered by attacker-controlled input, allowing a single crafted request to consume disproportionate CPU, memory, or backing-store resources. Because search endpoints typically translate request parameters into database or in-memory filtering operations, the likely root cause is an unbounded or overly expensive query/filter path with no input size, pagination, or complexity limits. The CPE data is non-informative (cpe:2.3:a:n/a:n/a), so exact affected-product identification relies on the CVE description and the GitHub repository reference rather than NVD CPE.
RemediationAI
No vendor-released patch version is identified from the available data; the references point to the project's main branch (https://github.com/ibbybuilds/aegra/tree/main) and a gist (https://gist.github.com/syphonetic/d27b5965c884555acea1827988689f7d), not a tagged patched release, so operators should track the upstream repository, compare their deployed commit against e9a89f, and update to a post-fix commit once confirmed. As compensating controls until a fix is confirmed, restrict network access to the /store/items/search endpoint by placing Aegra behind authentication or a reverse proxy and denying anonymous access (trade-off: breaks any client that legitimately calls it unauthenticated), enforce rate limiting and request-size/timeout caps at the proxy on the search endpoint to blunt resource exhaustion (trade-off: may throttle high-volume legitimate search traffic), and avoid exposing the server directly to the internet by binding it to internal networks or a VPN (trade-off: reduces convenience of remote access).
LibreNMS before 24.10.0 allows a remote attacker to execute arbitrary code via OS command injection involving AboutContr
Reachable assertion in Open5GS up to version 2.7.6 allows remote unauthenticated attackers to crash the Diameter CER han
A Command Injection issue in the payload build page in BYOB (Build Your Own Botnet) 2.0 allows attackers to execute arbi
Denial of service in Open5GS 2.7.6 and earlier allows network-adjacent attackers with low-privilege Diameter Gx peer acc
Denial of service in Open5GS HSS (Home Subscriber Server) versions 2.7.0 through 2.7.6 is triggered remotely by manipula
Remote command injection in Control Web Panel allows unauthenticated attackers to execute arbitrary OS commands as root
LINQPad before 5.52.01 Pro edition is vulnerable to Unsafe Deserialization in LINQPad.AutoRefManager::PopulateFromCache(
Remote code execution in dash-uploader (Python package for Plotly Dash) versions 0.1.0 through 0.7.0a2 allows unauthenti
Command injection in LiteLLM 1.18.10 lets attackers who can configure MCP (Model Context Protocol) servers supply arbitr
SQL injection in AzuraCast's backup restore flow (versions up to and including 0.23.7) lets an attacker achieve privileg
Arbitrary file read in Yu AI Code Mother v4.3 lets anonymous, remote attackers retrieve files outside the intended previ
Remote code execution in APScheduler (all versions through 3.10.x and 4.0.0a5) is achievable when applications deseriali
Same weakness CWE-400 – Uncontrolled Resource Consumption
View allSame technique Denial Of Service
View allShare
External POC / Exploit Code
Leaving vuln.today