mirror of
https://github.com/openai/codex.git
synced 2026-05-04 21:32:21 +03:00
CHAINED PR - note that base is eternal/hooks-pretooluse-bash, not main -- so the following PR should be first Matching post-tool hook to the pre-tool functionality here: https://github.com/openai/codex/pull/15211 So, PreToolUse calls for plain shell calls, allows blocking. This PostToolUse call runs after the command executed example run: ``` › as a test, run in parallel the following commands: - echo 'one' - echo '[block-pre-tool-use]' - echo '[block-post-tool-use]' ⚠ MCP startup incomplete (failed: notion, linear) • Cruising through those three commands in parallel now, and I’ll share the exact outputs right after they land. • Running PreToolUse hook: checking the observatory runes • Running PreToolUse hook: checking the observatory runes • Running PreToolUse hook: checking the observatory runes PreToolUse hook (blocked) warning: wizard-tower PreToolUse demo blocked a Bash command on purpose. feedback: PreToolUse demo blocked the command. Remove [block-pre-tool-use] to continue. PreToolUse hook (completed) warning: wizard-tower PreToolUse demo inspected Bash: echo 'one' PreToolUse hook (completed) warning: wizard-tower PreToolUse demo inspected Bash: echo '[block-post-tool-use]' • Ran echo '[block-post-tool-use]' └ [block-post-tool-use] • Ran echo 'one' └ one • Running PostToolUse hook: reading the observatory aftermath • Running PostToolUse hook: reading the observatory aftermath PostToolUse hook (blocked) warning: wizard-tower PostToolUse demo blocked the Bash follow-through on purpose. feedback: PostToolUse demo blocked the result after execution. Remove [block-post-tool-use] to continue. PostToolUse hook (completed) warning: wizard-tower PostToolUse demo inspected Bash: echo 'one' hook context: PostToolUse demo saw the Bash result. Response preview: one ────────────────────────────────────────────────────────────────────────────────────────────────────── • Ran them in parallel. Little wave report: - echo 'one' completed and returned one. - echo '[block-pre-tool-use]' was blocked before execution by the PreToolUse hook. - echo '[block-post-tool-use]' ran, but its result was blocked afterward by the PostToolUse hook. The hook messages were: - PreToolUse demo blocked the command. Remove [block-pre-tool-use] to continue. - PostToolUse demo blocked the result after execution. Remove [block-post-tool-use] to continue. › for the blocked messages (correctly blocked in this test) did the tool output land back into context? • Yeah, from this run: - PreToolUse block: the command itself did not execute, so no command output landed in context. Only the block message did. - PostToolUse block: the command did execute, but the actual stdout did not come back into context. Only the post-tool block message did. So the blocked notifications landed in context, but the blocked command payloads themselves did not. Nice clean guardrail, aloha. ```
381 lines
12 KiB
Rust
381 lines
12 KiB
Rust
use std::future::Future;
|
|
use std::sync::Arc;
|
|
|
|
use codex_hooks::PostToolUseOutcome;
|
|
use codex_hooks::PostToolUseRequest;
|
|
use codex_hooks::PreToolUseOutcome;
|
|
use codex_hooks::PreToolUseRequest;
|
|
use codex_hooks::SessionStartOutcome;
|
|
use codex_hooks::UserPromptSubmitOutcome;
|
|
use codex_hooks::UserPromptSubmitRequest;
|
|
use codex_protocol::items::TurnItem;
|
|
use codex_protocol::models::DeveloperInstructions;
|
|
use codex_protocol::models::ResponseInputItem;
|
|
use codex_protocol::models::ResponseItem;
|
|
use codex_protocol::protocol::AskForApproval;
|
|
use codex_protocol::protocol::EventMsg;
|
|
use codex_protocol::protocol::HookCompletedEvent;
|
|
use codex_protocol::protocol::HookRunSummary;
|
|
use codex_protocol::user_input::UserInput;
|
|
use serde_json::Value;
|
|
|
|
use crate::codex::Session;
|
|
use crate::codex::TurnContext;
|
|
use crate::event_mapping::parse_turn_item;
|
|
|
|
pub(crate) struct HookRuntimeOutcome {
|
|
pub should_stop: bool,
|
|
pub additional_contexts: Vec<String>,
|
|
}
|
|
|
|
pub(crate) enum PendingInputHookDisposition {
|
|
Accepted(Box<PendingInputRecord>),
|
|
Blocked { additional_contexts: Vec<String> },
|
|
}
|
|
|
|
pub(crate) enum PendingInputRecord {
|
|
UserMessage {
|
|
content: Vec<UserInput>,
|
|
response_item: ResponseItem,
|
|
additional_contexts: Vec<String>,
|
|
},
|
|
ConversationItem {
|
|
response_item: ResponseItem,
|
|
},
|
|
}
|
|
|
|
struct ContextInjectingHookOutcome {
|
|
hook_events: Vec<HookCompletedEvent>,
|
|
outcome: HookRuntimeOutcome,
|
|
}
|
|
|
|
impl From<SessionStartOutcome> for ContextInjectingHookOutcome {
|
|
fn from(value: SessionStartOutcome) -> Self {
|
|
let SessionStartOutcome {
|
|
hook_events,
|
|
should_stop,
|
|
stop_reason: _,
|
|
additional_contexts,
|
|
} = value;
|
|
Self {
|
|
hook_events,
|
|
outcome: HookRuntimeOutcome {
|
|
should_stop,
|
|
additional_contexts,
|
|
},
|
|
}
|
|
}
|
|
}
|
|
|
|
impl From<UserPromptSubmitOutcome> for ContextInjectingHookOutcome {
|
|
fn from(value: UserPromptSubmitOutcome) -> Self {
|
|
let UserPromptSubmitOutcome {
|
|
hook_events,
|
|
should_stop,
|
|
stop_reason: _,
|
|
additional_contexts,
|
|
} = value;
|
|
Self {
|
|
hook_events,
|
|
outcome: HookRuntimeOutcome {
|
|
should_stop,
|
|
additional_contexts,
|
|
},
|
|
}
|
|
}
|
|
}
|
|
|
|
pub(crate) async fn run_pending_session_start_hooks(
|
|
sess: &Arc<Session>,
|
|
turn_context: &Arc<TurnContext>,
|
|
) -> bool {
|
|
let Some(session_start_source) = sess.take_pending_session_start_source().await else {
|
|
return false;
|
|
};
|
|
|
|
let request = codex_hooks::SessionStartRequest {
|
|
session_id: sess.conversation_id,
|
|
cwd: turn_context.cwd.to_path_buf(),
|
|
transcript_path: sess.hook_transcript_path().await,
|
|
model: turn_context.model_info.slug.clone(),
|
|
permission_mode: hook_permission_mode(turn_context),
|
|
source: session_start_source,
|
|
};
|
|
let preview_runs = sess.hooks().preview_session_start(&request);
|
|
run_context_injecting_hook(
|
|
sess,
|
|
turn_context,
|
|
preview_runs,
|
|
sess.hooks()
|
|
.run_session_start(request, Some(turn_context.sub_id.clone())),
|
|
)
|
|
.await
|
|
.record_additional_contexts(sess, turn_context)
|
|
.await
|
|
}
|
|
|
|
pub(crate) async fn run_pre_tool_use_hooks(
|
|
sess: &Arc<Session>,
|
|
turn_context: &Arc<TurnContext>,
|
|
tool_use_id: String,
|
|
command: String,
|
|
) -> Option<String> {
|
|
let request = PreToolUseRequest {
|
|
session_id: sess.conversation_id,
|
|
turn_id: turn_context.sub_id.clone(),
|
|
cwd: turn_context.cwd.to_path_buf(),
|
|
transcript_path: sess.hook_transcript_path().await,
|
|
model: turn_context.model_info.slug.clone(),
|
|
permission_mode: hook_permission_mode(turn_context),
|
|
tool_name: "Bash".to_string(),
|
|
tool_use_id,
|
|
command,
|
|
};
|
|
let preview_runs = sess.hooks().preview_pre_tool_use(&request);
|
|
emit_hook_started_events(sess, turn_context, preview_runs).await;
|
|
|
|
let PreToolUseOutcome {
|
|
hook_events,
|
|
should_block,
|
|
block_reason,
|
|
} = sess.hooks().run_pre_tool_use(request).await;
|
|
emit_hook_completed_events(sess, turn_context, hook_events).await;
|
|
|
|
if should_block { block_reason } else { None }
|
|
}
|
|
|
|
pub(crate) async fn run_post_tool_use_hooks(
|
|
sess: &Arc<Session>,
|
|
turn_context: &Arc<TurnContext>,
|
|
tool_use_id: String,
|
|
command: String,
|
|
tool_response: Value,
|
|
) -> PostToolUseOutcome {
|
|
let request = PostToolUseRequest {
|
|
session_id: sess.conversation_id,
|
|
turn_id: turn_context.sub_id.clone(),
|
|
cwd: turn_context.cwd.to_path_buf(),
|
|
transcript_path: sess.hook_transcript_path().await,
|
|
model: turn_context.model_info.slug.clone(),
|
|
permission_mode: hook_permission_mode(turn_context),
|
|
tool_name: "Bash".to_string(),
|
|
tool_use_id,
|
|
command,
|
|
tool_response,
|
|
};
|
|
let preview_runs = sess.hooks().preview_post_tool_use(&request);
|
|
emit_hook_started_events(sess, turn_context, preview_runs).await;
|
|
|
|
let outcome = sess.hooks().run_post_tool_use(request).await;
|
|
emit_hook_completed_events(sess, turn_context, outcome.hook_events.clone()).await;
|
|
outcome
|
|
}
|
|
|
|
pub(crate) async fn run_user_prompt_submit_hooks(
|
|
sess: &Arc<Session>,
|
|
turn_context: &Arc<TurnContext>,
|
|
prompt: String,
|
|
) -> HookRuntimeOutcome {
|
|
let request = UserPromptSubmitRequest {
|
|
session_id: sess.conversation_id,
|
|
turn_id: turn_context.sub_id.clone(),
|
|
cwd: turn_context.cwd.to_path_buf(),
|
|
transcript_path: sess.hook_transcript_path().await,
|
|
model: turn_context.model_info.slug.clone(),
|
|
permission_mode: hook_permission_mode(turn_context),
|
|
prompt,
|
|
};
|
|
let preview_runs = sess.hooks().preview_user_prompt_submit(&request);
|
|
run_context_injecting_hook(
|
|
sess,
|
|
turn_context,
|
|
preview_runs,
|
|
sess.hooks().run_user_prompt_submit(request),
|
|
)
|
|
.await
|
|
}
|
|
|
|
pub(crate) async fn inspect_pending_input(
|
|
sess: &Arc<Session>,
|
|
turn_context: &Arc<TurnContext>,
|
|
pending_input_item: ResponseInputItem,
|
|
) -> PendingInputHookDisposition {
|
|
let response_item = ResponseItem::from(pending_input_item);
|
|
if let Some(TurnItem::UserMessage(user_message)) = parse_turn_item(&response_item) {
|
|
let user_prompt_submit_outcome =
|
|
run_user_prompt_submit_hooks(sess, turn_context, user_message.message()).await;
|
|
if user_prompt_submit_outcome.should_stop {
|
|
PendingInputHookDisposition::Blocked {
|
|
additional_contexts: user_prompt_submit_outcome.additional_contexts,
|
|
}
|
|
} else {
|
|
PendingInputHookDisposition::Accepted(Box::new(PendingInputRecord::UserMessage {
|
|
content: user_message.content,
|
|
response_item,
|
|
additional_contexts: user_prompt_submit_outcome.additional_contexts,
|
|
}))
|
|
}
|
|
} else {
|
|
PendingInputHookDisposition::Accepted(Box::new(PendingInputRecord::ConversationItem {
|
|
response_item,
|
|
}))
|
|
}
|
|
}
|
|
|
|
pub(crate) async fn record_pending_input(
|
|
sess: &Arc<Session>,
|
|
turn_context: &Arc<TurnContext>,
|
|
pending_input: PendingInputRecord,
|
|
) {
|
|
match pending_input {
|
|
PendingInputRecord::UserMessage {
|
|
content,
|
|
response_item,
|
|
additional_contexts,
|
|
} => {
|
|
sess.record_user_prompt_and_emit_turn_item(
|
|
turn_context.as_ref(),
|
|
content.as_slice(),
|
|
response_item,
|
|
)
|
|
.await;
|
|
record_additional_contexts(sess, turn_context, additional_contexts).await;
|
|
}
|
|
PendingInputRecord::ConversationItem { response_item } => {
|
|
sess.record_conversation_items(turn_context, std::slice::from_ref(&response_item))
|
|
.await;
|
|
}
|
|
}
|
|
}
|
|
|
|
async fn run_context_injecting_hook<Fut, Outcome>(
|
|
sess: &Arc<Session>,
|
|
turn_context: &Arc<TurnContext>,
|
|
preview_runs: Vec<HookRunSummary>,
|
|
outcome_future: Fut,
|
|
) -> HookRuntimeOutcome
|
|
where
|
|
Fut: Future<Output = Outcome>,
|
|
Outcome: Into<ContextInjectingHookOutcome>,
|
|
{
|
|
emit_hook_started_events(sess, turn_context, preview_runs).await;
|
|
|
|
let outcome = outcome_future.await.into();
|
|
emit_hook_completed_events(sess, turn_context, outcome.hook_events).await;
|
|
outcome.outcome
|
|
}
|
|
|
|
impl HookRuntimeOutcome {
|
|
async fn record_additional_contexts(
|
|
self,
|
|
sess: &Arc<Session>,
|
|
turn_context: &Arc<TurnContext>,
|
|
) -> bool {
|
|
record_additional_contexts(sess, turn_context, self.additional_contexts).await;
|
|
|
|
self.should_stop
|
|
}
|
|
}
|
|
|
|
pub(crate) async fn record_additional_contexts(
|
|
sess: &Arc<Session>,
|
|
turn_context: &Arc<TurnContext>,
|
|
additional_contexts: Vec<String>,
|
|
) {
|
|
let developer_messages = additional_context_messages(additional_contexts);
|
|
if developer_messages.is_empty() {
|
|
return;
|
|
}
|
|
|
|
sess.record_conversation_items(turn_context, developer_messages.as_slice())
|
|
.await;
|
|
}
|
|
|
|
fn additional_context_messages(additional_contexts: Vec<String>) -> Vec<ResponseItem> {
|
|
additional_contexts
|
|
.into_iter()
|
|
.map(|additional_context| DeveloperInstructions::new(additional_context).into())
|
|
.collect()
|
|
}
|
|
|
|
async fn emit_hook_started_events(
|
|
sess: &Arc<Session>,
|
|
turn_context: &Arc<TurnContext>,
|
|
preview_runs: Vec<HookRunSummary>,
|
|
) {
|
|
for run in preview_runs {
|
|
sess.send_event(
|
|
turn_context,
|
|
EventMsg::HookStarted(crate::protocol::HookStartedEvent {
|
|
turn_id: Some(turn_context.sub_id.clone()),
|
|
run,
|
|
}),
|
|
)
|
|
.await;
|
|
}
|
|
}
|
|
|
|
async fn emit_hook_completed_events(
|
|
sess: &Arc<Session>,
|
|
turn_context: &Arc<TurnContext>,
|
|
completed_events: Vec<HookCompletedEvent>,
|
|
) {
|
|
for completed in completed_events {
|
|
sess.send_event(turn_context, EventMsg::HookCompleted(completed))
|
|
.await;
|
|
}
|
|
}
|
|
|
|
fn hook_permission_mode(turn_context: &TurnContext) -> String {
|
|
match turn_context.approval_policy.value() {
|
|
AskForApproval::Never => "bypassPermissions",
|
|
AskForApproval::UnlessTrusted
|
|
| AskForApproval::OnFailure
|
|
| AskForApproval::OnRequest
|
|
| AskForApproval::Granular(_) => "default",
|
|
}
|
|
.to_string()
|
|
}
|
|
|
|
#[cfg(test)]
|
|
mod tests {
|
|
use codex_protocol::models::ContentItem;
|
|
use pretty_assertions::assert_eq;
|
|
|
|
use super::additional_context_messages;
|
|
|
|
#[test]
|
|
fn additional_context_messages_stay_separate_and_ordered() {
|
|
let messages = additional_context_messages(vec![
|
|
"first tide note".to_string(),
|
|
"second tide note".to_string(),
|
|
]);
|
|
|
|
assert_eq!(messages.len(), 2);
|
|
assert_eq!(
|
|
messages
|
|
.iter()
|
|
.map(|message| match message {
|
|
codex_protocol::models::ResponseItem::Message { role, content, .. } => {
|
|
let text = content
|
|
.iter()
|
|
.map(|item| match item {
|
|
ContentItem::InputText { text } => text.as_str(),
|
|
ContentItem::InputImage { .. } | ContentItem::OutputText { .. } => {
|
|
panic!("expected input text content, got {item:?}")
|
|
}
|
|
})
|
|
.collect::<String>();
|
|
(role.as_str(), text)
|
|
}
|
|
other => panic!("expected developer message, got {other:?}"),
|
|
})
|
|
.collect::<Vec<_>>(),
|
|
vec![
|
|
("developer", "first tide note".to_string()),
|
|
("developer", "second tide note".to_string()),
|
|
],
|
|
);
|
|
}
|
|
}
|