Grimoire EUVD-2026-114114

Exploitation requires network reachability to the Grimoire /api/fetch-metadata POST handler, which per the description performs the fetch without any authentication check - so no credentials, user interaction, or non-default setting are needed (PR:N/UI:N confirms this). The concrete prerequisite is simply that the Grimoire instance (before 1.0.0) is network-accessible to the attacker; the attacker supplies a URL targeting a loopback, internal-network, or cloud metadata endpoint. Limiting factors: impact is read-only disclosure of the parsed title, description, and page content returned by getMetadata() (VC:L, no integrity or availability impact), and real escalation depends on whether reachable internal services (e.g. a cloud metadata service) expose sensitive data to an unauthenticated GET-style fetch. Instances not exposed to the attacker's network segment are out of reach.

The signals point to a moderate, not emergency, priority. The CVSS 4.0 vector (AV:N/AC:L/AT:N/PR:N/UI:N) confirms remote, unauthenticated, low-complexity reachability with no user interaction, which is the worst-case access profile; however impact is bounded (VC:L/VI:N/VA:N, SC:L) - the attacker reads back parsed title, description, and page content from the forged request rather than achieving write, RCE, or full data compromise, yielding the 6.9 MEDIUM base score. The vuln.today decision label is 'this_month' (not act_now/emergency), consistent with meaningful but non-critical urgency. No CISA KEV listing is present in the input, so there is no confirmation of active exploitation, and no EPSS score was provided, so statistical exploitation probability cannot be weighed here. A public PoC reference exists (a HackMD write-up and the upstream GitHub issue) but it is not verified as working exploit code. Net: a genuine, easily-reachable SSRF worth patching this cycle, but the limited (read-only, low-confidentiality) impact and absence of KEV/EPSS data argue against treating it as a drop-everything issue - the main escalation risk is using the SSRF to read cloud metadata (e.g. IMDS) credentials, which would elevate real-world severity beyond the base score.

An attacker sends an unauthenticated POST to /api/fetch-metadata with a URL pointing at a loopback, internal-network, or cloud metadata endpoint (e.g. 169.254.169.254), and the server fetches it via getMetadata() and returns the parsed title, description, and page content. A public PoC reference exists but is not verified as working exploit code. Because the vector is AV:N/AC:L/PR:N/UI:N, no authentication, user interaction, or special timing is required.