Files
codex/codex-rs/core/src/sandboxing/mod.rs
Eric Traut f8af4f5c8d Added model summary and risk assessment for commands that violate sandbox policy (#5536)
This PR adds support for a model-based summary and risk assessment for
commands that violate the sandbox policy and require user approval. This
aids the user in evaluating whether the command should be approved.

The feature works by taking a failed command and passing it back to the
model and asking it to summarize the command, give it a risk level (low,
medium, high) and a risk category (e.g. "data deletion" or "data
exfiltration"). It uses a new conversation thread so the context in the
existing thread doesn't influence the answer. If the call to the model
fails or takes longer than 5 seconds, it falls back to the current
behavior.

For now, this is an experimental feature and is gated by a config key
`experimental_sandbox_command_assessment`.

Here is a screen shot of the approval prompt showing the risk assessment
and summary.

<img width="723" height="282" alt="image"
src="https://github.com/user-attachments/assets/4597dd7c-d5a0-4e9f-9d13-414bd082fd6b"
/>
2025-10-24 15:23:44 -07:00

174 lines
5.5 KiB
Rust
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
/*
Module: sandboxing
Build platform wrappers and produce ExecEnv for execution. Owns lowlevel
sandbox placement and transformation of portable CommandSpec into a
readytospawn environment.
*/
pub mod assessment;
use crate::exec::ExecToolCallOutput;
use crate::exec::SandboxType;
use crate::exec::StdoutStream;
use crate::exec::execute_exec_env;
use crate::landlock::create_linux_sandbox_command_args;
use crate::protocol::SandboxPolicy;
use crate::seatbelt::MACOS_PATH_TO_SEATBELT_EXECUTABLE;
use crate::seatbelt::create_seatbelt_command_args;
use crate::spawn::CODEX_SANDBOX_ENV_VAR;
use crate::spawn::CODEX_SANDBOX_NETWORK_DISABLED_ENV_VAR;
use crate::tools::sandboxing::SandboxablePreference;
use std::collections::HashMap;
use std::path::Path;
use std::path::PathBuf;
#[derive(Clone, Debug)]
pub struct CommandSpec {
pub program: String,
pub args: Vec<String>,
pub cwd: PathBuf,
pub env: HashMap<String, String>,
pub timeout_ms: Option<u64>,
pub with_escalated_permissions: Option<bool>,
pub justification: Option<String>,
}
#[derive(Clone, Debug)]
pub struct ExecEnv {
pub command: Vec<String>,
pub cwd: PathBuf,
pub env: HashMap<String, String>,
pub timeout_ms: Option<u64>,
pub sandbox: SandboxType,
pub with_escalated_permissions: Option<bool>,
pub justification: Option<String>,
pub arg0: Option<String>,
}
pub enum SandboxPreference {
Auto,
Require,
Forbid,
}
#[derive(Debug, thiserror::Error)]
pub(crate) enum SandboxTransformError {
#[error("missing codex-linux-sandbox executable path")]
MissingLinuxSandboxExecutable,
}
#[derive(Default)]
pub struct SandboxManager;
impl SandboxManager {
pub fn new() -> Self {
Self
}
pub(crate) fn select_initial(
&self,
policy: &SandboxPolicy,
pref: SandboxablePreference,
) -> SandboxType {
match pref {
SandboxablePreference::Forbid => SandboxType::None,
SandboxablePreference::Require => {
#[cfg(target_os = "macos")]
{
return SandboxType::MacosSeatbelt;
}
#[cfg(target_os = "linux")]
{
return SandboxType::LinuxSeccomp;
}
#[allow(unreachable_code)]
SandboxType::None
}
SandboxablePreference::Auto => match policy {
SandboxPolicy::DangerFullAccess => SandboxType::None,
#[cfg(target_os = "macos")]
_ => SandboxType::MacosSeatbelt,
#[cfg(target_os = "linux")]
_ => SandboxType::LinuxSeccomp,
#[cfg(not(any(target_os = "macos", target_os = "linux")))]
_ => SandboxType::None,
},
}
}
pub(crate) fn transform(
&self,
spec: &CommandSpec,
policy: &SandboxPolicy,
sandbox: SandboxType,
sandbox_policy_cwd: &Path,
codex_linux_sandbox_exe: Option<&PathBuf>,
) -> Result<ExecEnv, SandboxTransformError> {
let mut env = spec.env.clone();
if !policy.has_full_network_access() {
env.insert(
CODEX_SANDBOX_NETWORK_DISABLED_ENV_VAR.to_string(),
"1".to_string(),
);
}
let mut command = Vec::with_capacity(1 + spec.args.len());
command.push(spec.program.clone());
command.extend(spec.args.iter().cloned());
let (command, sandbox_env, arg0_override) = match sandbox {
SandboxType::None => (command, HashMap::new(), None),
SandboxType::MacosSeatbelt => {
let mut seatbelt_env = HashMap::new();
seatbelt_env.insert(CODEX_SANDBOX_ENV_VAR.to_string(), "seatbelt".to_string());
let mut args =
create_seatbelt_command_args(command.clone(), policy, sandbox_policy_cwd);
let mut full_command = Vec::with_capacity(1 + args.len());
full_command.push(MACOS_PATH_TO_SEATBELT_EXECUTABLE.to_string());
full_command.append(&mut args);
(full_command, seatbelt_env, None)
}
SandboxType::LinuxSeccomp => {
let exe = codex_linux_sandbox_exe
.ok_or(SandboxTransformError::MissingLinuxSandboxExecutable)?;
let mut args =
create_linux_sandbox_command_args(command.clone(), policy, sandbox_policy_cwd);
let mut full_command = Vec::with_capacity(1 + args.len());
full_command.push(exe.to_string_lossy().to_string());
full_command.append(&mut args);
(
full_command,
HashMap::new(),
Some("codex-linux-sandbox".to_string()),
)
}
};
env.extend(sandbox_env);
Ok(ExecEnv {
command,
cwd: spec.cwd.clone(),
env,
timeout_ms: spec.timeout_ms,
sandbox,
with_escalated_permissions: spec.with_escalated_permissions,
justification: spec.justification.clone(),
arg0: arg0_override,
})
}
pub fn denied(&self, sandbox: SandboxType, out: &ExecToolCallOutput) -> bool {
crate::exec::is_likely_sandbox_denied(sandbox, out)
}
}
pub async fn execute_env(
env: &ExecEnv,
policy: &SandboxPolicy,
stdout_stream: Option<StdoutStream>,
) -> crate::error::Result<ExecToolCallOutput> {
execute_exec_env(env.clone(), policy, stdout_stream).await
}