CVE-2026-57120: PraisonAI Execute_code Sandbox Bypass: Str.format C-level Vulnerability

{
  "schema_version" : "1.4.0",
  "id" : "GHSA-pv2j-rghr-v5r9",
  "modified" : "2026-07-20T21:24:25Z",
  "published" : "2026-06-18T13:55:29Z",
  "aliases" : [ "CVE-2026-57120" ],
  "summary" : "PraisonAI: execute_code sandbox bypass: str.format C-level attribute access reads every blocklisted dunder",
  "details" : "## Summary\n\nThe `execute_code` tool's subprocess sandbox advertises a three-layer defense (AST validation, text-pattern blocklist, restricted `__builtins__`). In **sandbox mode** (the default) only two layers are active — the text-pattern blocklist is skipped — and both remaining layers are bypassed by combining two CPython semantics:\n\n1. **Runtime string assembly.** The AST validator (`src/praisonai-agents/praisonaiagents/tools/python_tools.py:75`) enumerates blocked dunder names against `ast.Attribute.attr`, `ast.Call.func.id`, and `ast.Constant` string-substring. Names assembled at runtime (e.g. `\"_\"*2 + \"class\" + \"_\"*2`) appear in the AST as multiple short `ast.Constant` nodes, none containing a blocked substring, so the static check passes.\n2. **C-level attribute access via format-spec.** `str.format` / `str.format_map` resolve dotted field references through CPython's internal `PyObject_GetAttr` (`do_string_format` → `get_field`). This C path never consults the Python-level `getattr` binding. The sandbox's `_safe_getattr` wrapper (`python_tools.py:221`) is installed only as the `getattr` name in `safe_builtins`, so any C-level attribute access — including format-spec field resolution — sidesteps it. `format`/`format_map` are also absent from `_SANDBOX_BLOCKED_CALLS` (`python_tools.py:56`).\n\nCombined, this yields an arbitrary read primitive over every blocklisted attribute (`__class__`, `__qualname__`, `__bases__`, `__base__`, function `__globals__`, `__dict__`, …).\n\n## Affected code\n\n| File | Lines | Symbol | Role |\n|---|---|---|---|\n| `src/praisonai-agents/praisonaiagents/tools/python_tools.py` | 39–54 | `_SANDBOX_BLOCKED_ATTRS` | The blocklist the bypass renders unreachable |\n| `src/praisonai-agents/praisonaiagents/tools/python_tools.py` | 56–60 | `_SANDBOX_BLOCKED_CALLS` | Missing entries: `format`, `format_map` |\n| `src/praisonai-agents/praisonaiagents/tools/python_tools.py` | 75–102 | `_validate_code_ast` | Static check, blind to runtime string assembly |\n| `src/praisonai-agents/praisonaiagents/tools/python_tools.py` | 221–226 | `_safe_getattr` | Wraps Python-level `getattr` only; C-level access bypasses |\n| `src/praisonai-agents/praisonaiagents/tools/python_tools.py` | 352 | `execute_code` | Entry point; gated by `@require_approval(risk_level=\"critical\")` |\n\n## Reproducer\n\n```python\nimport os\nos.environ[\"PRAISONAI_AUTO_APPROVE\"] = \"true\"\nfrom praisonaiagents.tools.python_tools import execute_code\n\npayload = '''\nund = \"_\" * 2                         # \"__\" assembled at runtime\nkey1 = und + \"class\" + und            # \"__class__\"\nkey2 = und + \"qualname\" + und         # \"__qualname__\"\nfmt_class = \"{0.\" + key1 + \"}\"\nfmt_qual2 = \"{0.\" + key1 + \".\" + key2 + \"}\"\nprint(\"LEAK_CLASS=\" + fmt_class.format(()))\nprint(\"LEAK_QUAL2=\" + fmt_qual2.format(()))\n'''\nprint(execute_code(payload, sandbox_mode=\"sandbox\", timeout=15))\n```\n\nObserved: `stdout` = `LEAK_CLASS=<class 'tuple'>` / `LEAK_QUAL2=tuple`, `success: true`, no security error. Both `__class__` (one hop) and `__class__.__qualname__` (two hops) — all blocklisted — are read.\n\n## Trust boundary\n\nThe `@require_approval(risk_level=\"critical\")` gate is bypassed when `PRAISONAI_AUTO_APPROVE` is set (verified: `require_approval` short-circuits on `is_env_auto_approve()`). That variable is set by the project's FULL_AUTO autonomy mode, the bots-CLI launchers, and the project's own issue-triage CI workflow — postures where the agent reaches `execute_code` with no human approval. The payload then arrives via any LLM-visible surface (user message, retrieved document, tool/web/MCP output) and the tool-call machinery passes it as the `code` argument.\n\n## Relationship to GHSA-4mr5-g6f9-cfrh\n\nThe code's own comment at `python_tools.py:46` cites GHSA-4mr5-g6f9-cfrh, which added `__self__` to the blocklist to stop C-builtins leaking `builtins` via `func.__self__`. This finding does not bypass that single entry — it bypasses the **entire** blocklist, because format-spec attribute resolution never consults the blocklist or `_safe_getattr`. `\"{0.__self__}\".format(print)` would leak `__self__` regardless of the blocklist. Same defense surface, different mechanism; the GHSA-4mr5 fix does not mitigate this.\n\n## Scope (read primitive only)\n\nThis reports the **read primitive**. Turning the read into in-process execution requires a callable bridge; the obvious one (`string.Formatter().get_field()` returning the live object) is not directly reachable because `import string` is blocked at the AST layer (no `ast.Import`). Other bridges may exist; a full execution chain is **not** claimed here. If one is found, severity rises to ~8.8 (the subprocess has no seccomp/`setrlimit`/syscall filtering).\n\n## Suggested fix\n\n1. Add `format`, `format_map` to `_SANDBOX_BLOCKED_CALLS` (blocks the calls at the AST layer; cost: also blocks benign `str.format`).\n2. Or replace `str` in `safe_builtins` with a subclass whose `format`/`format_map` reject dotted fields resolving to leading-underscore attributes (preserves benign formatting).\n3. Or drop sandbox-mode's in-process security claim and document that real isolation requires external sandboxing (gVisor/firejail/container/microVM) — which matches what the subprocess provides today.\n\nThe text-pattern blocklist present in the `direct` path (`python_tools.py:487-502`) is absent from the sandbox path; even if added, the runtime-assembly trick defeats it, so (1) or (2) is required.\n\nReporter: Kai Aizen / SnailSploit — kai@snailsploit.com — PGP on request. Coordinated disclosure; no public posting.",
  "severity" : [ {
    "type" : "CVSS_V3",
    "score" : "CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N"
  } ],
  "affected" : [ {
    "package" : {
      "ecosystem" : "PyPI",
      "name" : "praisonaiagents"
    },
    "ranges" : [ {
      "type" : "ECOSYSTEM",
      "events" : [ {
        "introduced" : "0"
      }, {
        "fixed" : "1.6.59"
      } ]
    } ]
  } ],
  "references" : [ {
    "type" : "WEB",
    "url" : "https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-pv2j-rghr-v5r9"
  }, {
    "type" : "PACKAGE",
    "url" : "https://github.com/MervinPraison/PraisonAI"
  } ],
  "database_specific" : {
    "cwe_ids" : [ "CWE-693" ],
    "severity" : "MODERATE",
    "github_reviewed" : true,
    "github_reviewed_at" : "2026-06-18T13:55:29Z",
    "nvd_published_at" : null
  }
}