MISP SSRF and Credential Forwarding via Insufficient Outbound URL Validation

Technical Analysis

The vulnerability stems from two related flaws: (1) MISP's feed processing followed HTTP redirects without validating the redirect scheme or destination, and reused original request headers—including feed authentication credentials or API keys—across redirect hops to different hosts, enabling credential exfiltration and SSRF into internal networks; (2) the TAXII discovery endpoint relied on gethostbyname() checked against a small blocklist of literal addresses, missing bypasses via IPv6 loopback (::1), numeric/hex IP encodings (e.g., 0x7f000001), and multi-record DNS responses, allowing DNS rebinding-style attacks. The fix introduces a shared URL egress validator that blocks internal destinations on cross-host redirects, strips credentials before following external redirects, and pins DNS resolution results to prevent re-resolution after validation (mitigating TOCTOU/DNS rebinding). Exploitation requires an attacker to control or influence a feed/TAXII source configured in a victim's MISP instance, then use redirects to pivot into internal services (cloud metadata endpoints, internal APIs, admin panels) or steal feed credentials. For organizations using MISP as a threat-intel source feeding AI-driven SOAR/agent pipelines or RAG-based security assistants, a successful SSRF or credential leak could poison ingested threat data, expose API keys used by downstream automation, or grant lateral network access to systems hosting agent orchestration components.

Affected Systems

MISP (Malware Information Sharing Platform) versions ≤2.5.45; specifically the feed retrieval subsystem and TAXII discovery endpoint; instances configured with authenticated external feeds or TAXII servers

Indicators of Compromise

• No specific file hashes, IPs, or domains published; indicators would be environment-specific (malicious feed URLs, unexpected redirect chains, anomalous internal network requests originating from MISP server, unexpected DNS resolution changes for configured feed hostnames)

Remediation Steps

1. 1

   Upgrade MISP

   Update to the patched MISP release that includes the shared URL egress validator, redirect scheme/destination validation, DNS pinning, and credential stripping on cross-host redirects.

2. 2

   Audit configured feeds and TAXII sources

   Review all feed and TAXII server configurations for legitimacy, remove untrusted or unverified sources, and rotate any credentials associated with external feeds.

3. 3

   Implement network egress controls

   Restrict outbound HTTP requests from the MISP host via firewall/proxy rules to prevent access to internal/private IP ranges and cloud metadata endpoints, independent of application-layer fixes.

4. 4

   Rotate exposed credentials

   Rotate any API keys or authentication headers used in feed configurations, since these may have been forwarded to unintended hosts prior to patching.

5. 5

   Monitor outbound traffic

   Enable logging and alerting on outbound connections from the MISP server to detect anomalous destinations or repeated redirect chains indicative of exploitation attempts.

6. 6

   Review downstream integrations

   For organizations feeding MISP threat intel into automated agents, SOAR platforms, or RAG pipelines, verify data integrity and re-validate any credentials shared between MISP and those systems.