CVE-2026-59726: Ruflo: Unauthenticated RCE in MCP bridge default docker-compose deployment
Ruflo is an agent meta-harness for Claude Code and Codex. Prior to 3.16.3, ruflo's default docker-compose deployment exposed the MCP bridge POST /mcp and POST /mcp/:group endpoints without authentication, allowing an unauthenticated network attacker to invoke tools/call to terminal_execute, obtain a shell in the bridge container, read provider API keys, and poison AgentDB learning-store patterns. This issue is fixed in version 3.16.3.
Security readout for executives and security teams
Plain-English summary
CVE-2026-59726 is a critical issue in Ruflo before 3.16.3. The default docker-compose deployment exposed MCP bridge endpoints without authentication. A network attacker could run terminal actions inside the bridge container, access provider API keys, and alter AgentDB learning-store patterns. This is fixed in Ruflo 3.16.3.
Executive priority
Treat as an immediate remediation item for any exposed Ruflo deployment. The business risk is secret theft, attacker command execution inside the bridge container, and manipulation of agent behavior. Prioritize upgrade and exposure reduction before broader hardening.
Technical view
Ruflo versions before 3.16.3 expose POST /mcp and POST /mcp/:group unauthenticated in the default docker-compose MCP bridge. The advisory describes unauthenticated tool invocation leading to terminal_execute, container shell access, credential exposure, and AgentDB poisoning. CVSS 3.1 is 10.0 with network, low-complexity, no-privilege, no-user-interaction characteristics.
Likely exposure
Highest concern is Ruflo deployments using the default docker-compose MCP bridge and reachable over an untrusted network. Exposure depends on whether the MCP bridge endpoints are network-accessible and whether versions are below 3.16.3.
Exploitation context
The CVE record and GitHub advisory describe unauthenticated remote code execution potential. The provided data does not show CISA KEV listing or confirmed active exploitation. The issue is still urgent because no credentials or user interaction are required when the vulnerable service is exposed.
Researcher notes
Evidence comes from the CVE record and Ruflo GitHub advisory, PR, commit, and v3.16.3 release. The sources name the affected product and fixed version, but do not provide independent exploitation telemetry. Avoid assuming impact beyond the MCP bridge container and described credential or AgentDB effects.
Mitigation direction
Upgrade Ruflo to version 3.16.3 or later.
Restrict network access to MCP bridge endpoints.
Remove public exposure of default docker-compose MCP bridge deployments.
Rotate provider API keys if exposure is suspected.
Review vendor advisory and release notes for configuration guidance.
Validation and detection
Inventory Ruflo deployments and identify versions below 3.16.3.
Check whether /mcp or /mcp/:group are reachable from untrusted networks.
Review docker-compose deployments for default MCP bridge exposure.
Inspect logs for unexpected MCP tool calls or terminal activity.
Check AgentDB learning-store patterns for unauthorized changes.
Generated from the cited source records. This long-tail analysis has not been individually reviewed by a named human.
Potential ATT&CK relevance
Conservative CVE-to-ATT&CK context
These mappings and lookup hints may be relevant to the vulnerability behavior, CWE, affected product, or exposure path. Glexia-inferred context is not an official MITRE, ATT&CK, CWE, or CVE Program mapping.
ATT&CK lookup starting points
Use these exact CWE pages and searches to review the Glexia ATT&CK library from this CVE's weakness and description context.
cwe · medium confidence lookup
CWE-306: Credential and account abuse lookup
Authentication and credential weaknesses can make valid-account abuse and credential telemetry useful review starting points. Open the exact CWE lookup page first, then review the ATT&CK searches from that MITRE weakness context. This is a Glexia lookup hint, not an official ATT&CK mapping.
Command injection weaknesses can lead defenders to review execution techniques and command interpreter telemetry. Open the exact CWE lookup page first, then review the ATT&CK searches from that MITRE weakness context. This is a Glexia lookup hint, not an official ATT&CK mapping.
Use the exact CWE identifier as the starting point before reviewing related ATT&CK behavior. Open the exact CWE lookup page first, then review the ATT&CK searches from that MITRE weakness context. This is a Glexia lookup hint, not an official ATT&CK mapping.
The CVE wording references code or command execution, so execution technique review may help defensive triage. This is a Glexia inferred lookup path, not an official MITRE, ATT&CK, or CVE Program mapping.
The affected technology mentions containers, so container-specific ATT&CK technique review may help. This is a Glexia inferred lookup path, not an official MITRE, ATT&CK, or CVE Program mapping.
These fields come from the CVE record and ADP containers, not from Glexia's Take. They preserve time-varying source decisions such as CISA SSVC, KEV status, CVSS metrics, and provider references.
1CVSS vectors
3Timeline events
1ADP providers
5Source links
SSVC decision data
CISA-ADPCISA Coordinator
Timestamp
Version
2.0.3
Exploitation: noneAutomatable: yesTechnical Impact: total
CVSS vector scores
1 official score
We collect every scored CVSS vector available in the official CNA and ADP containers. When more than one version is present, the table keeps the source vectors side by side instead of collapsing them into the highest score.