[apply_patch] freeform apply_patch tool (#2576)

## Summary
GPT-5 introduced the concept of [custom
tools](https://platform.openai.com/docs/guides/function-calling#custom-tools),
which allow the model to send a raw string result back, simplifying
json-escape issues. We are migrating gpt-5 to use this by default.

However, gpt-oss models do not support custom tools, only normal
functions. So we keep both tool definitions, and provide whichever one
the model family supports.

## Testing
- [x] Tested locally with various models
- [x] Unit tests pass
This commit is contained in:
Dylan
2025-08-22 13:42:34 -07:00
committed by GitHub
parent dc42ec0eb4
commit 236c4f76a6
14 changed files with 534 additions and 138 deletions

View File

@@ -102,6 +102,33 @@ pub(crate) async fn stream_chat_completions(
"content": output.content,
}));
}
ResponseItem::CustomToolCall {
id,
call_id: _,
name,
input,
status: _,
} => {
messages.push(json!({
"role": "assistant",
"content": null,
"tool_calls": [{
"id": id,
"type": "custom",
"custom": {
"name": name,
"input": input,
}
}]
}));
}
ResponseItem::CustomToolCallOutput { call_id, output } => {
messages.push(json!({
"role": "tool",
"tool_call_id": call_id,
"content": output,
}));
}
ResponseItem::Reasoning { .. } | ResponseItem::Other => {
// Omit these items from the conversation history.
continue;

View File

@@ -575,6 +575,8 @@ async fn process_sse<S>(
}
"response.content_part.done"
| "response.function_call_arguments.delta"
| "response.custom_tool_call_input.delta"
| "response.custom_tool_call_input.done" // also emitted as response.output_item.done
| "response.in_progress"
| "response.output_item.added"
| "response.output_text.done" => {

View File

@@ -49,13 +49,14 @@ impl Prompt {
.unwrap_or(BASE_INSTRUCTIONS);
let mut sections: Vec<&str> = vec![base];
// When there are no custom instructions, add apply_patch if either:
// - the model needs special instructions, or
// When there are no custom instructions, add apply_patch_tool_instructions if either:
// - the model needs special instructions (4.1), or
// - there is no apply_patch tool present
let is_apply_patch_tool_present = self
.tools
.iter()
.any(|t| matches!(t, OpenAiTool::Function(f) if f.name == "apply_patch"));
let is_apply_patch_tool_present = self.tools.iter().any(|tool| match tool {
OpenAiTool::Function(f) => f.name == "apply_patch",
OpenAiTool::Freeform(f) => f.name == "apply_patch",
_ => false,
});
if self.base_instructions_override.is_none()
&& (model.needs_special_apply_patch_instructions || !is_apply_patch_tool_present)
{

View File

@@ -1406,6 +1406,18 @@ async fn run_task(
},
);
}
(
ResponseItem::CustomToolCall { .. },
Some(ResponseInputItem::CustomToolCallOutput { call_id, output }),
) => {
items_to_record_in_conversation_history.push(item);
items_to_record_in_conversation_history.push(
ResponseItem::CustomToolCallOutput {
call_id: call_id.clone(),
output: output.clone(),
},
);
}
(
ResponseItem::FunctionCall { .. },
Some(ResponseInputItem::McpToolCallOutput { call_id, result }),
@@ -1586,6 +1598,7 @@ async fn try_run_turn(
call_id: Some(call_id),
..
} => Some(call_id),
ResponseItem::CustomToolCallOutput { call_id, .. } => Some(call_id),
_ => None,
})
.collect::<Vec<_>>();
@@ -1603,6 +1616,7 @@ async fn try_run_turn(
call_id: Some(call_id),
..
} => Some(call_id),
ResponseItem::CustomToolCall { call_id, .. } => Some(call_id),
_ => None,
})
.filter_map(|call_id| {
@@ -1612,12 +1626,9 @@ async fn try_run_turn(
Some(call_id.clone())
}
})
.map(|call_id| ResponseItem::FunctionCallOutput {
.map(|call_id| ResponseItem::CustomToolCallOutput {
call_id: call_id.clone(),
output: FunctionCallOutputPayload {
content: "aborted".to_string(),
success: Some(false),
},
output: "aborted".to_string(),
})
.collect::<Vec<_>>()
};
@@ -1882,7 +1893,7 @@ async fn handle_response_item(
call_id,
..
} => {
info!("FunctionCall: {arguments}");
info!("FunctionCall: {name}({arguments})");
Some(
handle_function_call(
sess,
@@ -1939,10 +1950,32 @@ async fn handle_response_item(
.await,
)
}
ResponseItem::CustomToolCall {
id: _,
call_id,
name,
input,
status: _,
} => Some(
handle_custom_tool_call(
sess,
turn_context,
turn_diff_tracker,
sub_id.to_string(),
name,
input,
call_id,
)
.await,
),
ResponseItem::FunctionCallOutput { .. } => {
debug!("unexpected FunctionCallOutput from stream");
None
}
ResponseItem::CustomToolCallOutput { .. } => {
debug!("unexpected CustomToolCallOutput from stream");
None
}
ResponseItem::Other => None,
};
Ok(output)
@@ -2032,6 +2065,58 @@ async fn handle_function_call(
}
}
async fn handle_custom_tool_call(
sess: &Session,
turn_context: &TurnContext,
turn_diff_tracker: &mut TurnDiffTracker,
sub_id: String,
name: String,
input: String,
call_id: String,
) -> ResponseInputItem {
info!("CustomToolCall: {name} {input}");
match name.as_str() {
"apply_patch" => {
let exec_params = ExecParams {
command: vec!["apply_patch".to_string(), input.clone()],
cwd: turn_context.cwd.clone(),
timeout_ms: None,
env: HashMap::new(),
with_escalated_permissions: None,
justification: None,
};
let resp = handle_container_exec_with_params(
exec_params,
sess,
turn_context,
turn_diff_tracker,
sub_id,
call_id,
)
.await;
// Convert function-call style output into a custom tool call output
match resp {
ResponseInputItem::FunctionCallOutput { call_id, output } => {
ResponseInputItem::CustomToolCallOutput {
call_id,
output: output.content,
}
}
// Pass through if already a custom tool output or other variant
other => other,
}
}
_ => {
debug!("unexpected CustomToolCall from stream");
ResponseInputItem::CustomToolCallOutput {
call_id,
output: format!("unsupported custom tool call: {name}"),
}
}
}
}
fn to_exec_params(params: ShellToolCallParams, turn_context: &TurnContext) -> ExecParams {
ExecParams {
command: params.command,

View File

@@ -646,7 +646,7 @@ impl Config {
needs_special_apply_patch_instructions: false,
supports_reasoning_summaries,
uses_local_shell_tool: false,
uses_apply_patch_tool: false,
apply_patch_tool_type: None,
}
});
@@ -673,9 +673,6 @@ impl Config {
Self::get_base_instructions(experimental_instructions_path, &resolved_cwd)?;
let base_instructions = base_instructions.or(file_base_instructions);
let include_apply_patch_tool_val =
include_apply_patch_tool.unwrap_or(model_family.uses_apply_patch_tool);
let responses_originator_header: String = cfg
.responses_originator_header_internal_override
.unwrap_or(DEFAULT_RESPONSES_ORIGINATOR_HEADER.to_owned());
@@ -732,7 +729,7 @@ impl Config {
experimental_resume,
include_plan_tool: include_plan_tool.unwrap_or(false),
include_apply_patch_tool: include_apply_patch_tool_val,
include_apply_patch_tool: include_apply_patch_tool.unwrap_or(false),
responses_originator_header,
preferred_auth_method: cfg.preferred_auth_method.unwrap_or(AuthMode::ChatGPT),
};

View File

@@ -110,6 +110,8 @@ fn is_api_message(message: &ResponseItem) -> bool {
ResponseItem::Message { role, .. } => role.as_str() != "system",
ResponseItem::FunctionCallOutput { .. }
| ResponseItem::FunctionCall { .. }
| ResponseItem::CustomToolCall { .. }
| ResponseItem::CustomToolCallOutput { .. }
| ResponseItem::LocalShellCall { .. }
| ResponseItem::Reasoning { .. } => true,
ResponseItem::Other => false,

View File

@@ -50,6 +50,7 @@ pub mod seatbelt;
pub mod shell;
pub mod spawn;
pub mod terminal;
mod tool_apply_patch;
pub mod turn_diff_tracker;
pub mod user_agent;
mod user_notification;

View File

@@ -1,3 +1,5 @@
use crate::tool_apply_patch::ApplyPatchToolType;
/// A model family is a group of models that share certain characteristics.
#[derive(Debug, Clone, PartialEq, Eq, Hash)]
pub struct ModelFamily {
@@ -24,9 +26,9 @@ pub struct ModelFamily {
// See https://platform.openai.com/docs/guides/tools-local-shell
pub uses_local_shell_tool: bool,
/// True if the model performs better when `apply_patch` is provided as
/// a tool call instead of just a bash command.
pub uses_apply_patch_tool: bool,
/// Present if the model performs better when `apply_patch` is provided as
/// a tool call instead of just a bash command
pub apply_patch_tool_type: Option<ApplyPatchToolType>,
}
macro_rules! model_family {
@@ -40,7 +42,7 @@ macro_rules! model_family {
needs_special_apply_patch_instructions: false,
supports_reasoning_summaries: false,
uses_local_shell_tool: false,
uses_apply_patch_tool: false,
apply_patch_tool_type: None,
};
// apply overrides
$(
@@ -60,7 +62,7 @@ macro_rules! simple_model_family {
needs_special_apply_patch_instructions: false,
supports_reasoning_summaries: false,
uses_local_shell_tool: false,
uses_apply_patch_tool: false,
apply_patch_tool_type: None,
})
}};
}
@@ -88,6 +90,7 @@ pub fn find_family_for_model(slug: &str) -> Option<ModelFamily> {
model_family!(
slug, slug,
supports_reasoning_summaries: true,
apply_patch_tool_type: Some(ApplyPatchToolType::Freeform),
)
} else if slug.starts_with("gpt-4.1") {
model_family!(
@@ -95,7 +98,7 @@ pub fn find_family_for_model(slug: &str) -> Option<ModelFamily> {
needs_special_apply_patch_instructions: true,
)
} else if slug.starts_with("gpt-oss") {
model_family!(slug, "gpt-oss", uses_apply_patch_tool: true)
model_family!(slug, "gpt-oss", apply_patch_tool_type: Some(ApplyPatchToolType::Function))
} else if slug.starts_with("gpt-4o") {
simple_model_family!(slug, "gpt-4o")
} else if slug.starts_with("gpt-3.5") {
@@ -104,6 +107,7 @@ pub fn find_family_for_model(slug: &str) -> Option<ModelFamily> {
model_family!(
slug, "gpt-5",
supports_reasoning_summaries: true,
apply_patch_tool_type: Some(ApplyPatchToolType::Freeform),
)
} else {
None

View File

@@ -24,6 +24,10 @@ pub enum ResponseInputItem {
call_id: String,
result: Result<CallToolResult, String>,
},
CustomToolCallOutput {
call_id: String,
output: String,
},
}
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq)]
@@ -77,6 +81,20 @@ pub enum ResponseItem {
call_id: String,
output: FunctionCallOutputPayload,
},
CustomToolCall {
#[serde(default, skip_serializing_if = "Option::is_none")]
id: Option<String>,
#[serde(default, skip_serializing_if = "Option::is_none")]
status: Option<String>,
call_id: String,
name: String,
input: String,
},
CustomToolCallOutput {
call_id: String,
output: String,
},
#[serde(other)]
Other,
}
@@ -114,6 +132,9 @@ impl From<ResponseInputItem> for ResponseItem {
),
},
},
ResponseInputItem::CustomToolCallOutput { call_id, output } => {
Self::CustomToolCallOutput { call_id, output }
}
}
}
}

View File

@@ -9,6 +9,9 @@ use crate::model_family::ModelFamily;
use crate::plan_tool::PLAN_TOOL;
use crate::protocol::AskForApproval;
use crate::protocol::SandboxPolicy;
use crate::tool_apply_patch::ApplyPatchToolType;
use crate::tool_apply_patch::create_apply_patch_freeform_tool;
use crate::tool_apply_patch::create_apply_patch_json_tool;
#[derive(Debug, Clone, Serialize, PartialEq)]
pub struct ResponsesApiTool {
@@ -21,6 +24,20 @@ pub struct ResponsesApiTool {
pub(crate) parameters: JsonSchema,
}
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq)]
pub struct FreeformTool {
pub(crate) name: String,
pub(crate) description: String,
pub(crate) format: FreeformToolFormat,
}
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq)]
pub struct FreeformToolFormat {
pub(crate) r#type: String,
pub(crate) syntax: String,
pub(crate) definition: String,
}
/// When serialized as JSON, this produces a valid "Tool" in the OpenAI
/// Responses API.
#[derive(Debug, Clone, Serialize, PartialEq)]
@@ -30,6 +47,8 @@ pub(crate) enum OpenAiTool {
Function(ResponsesApiTool),
#[serde(rename = "local_shell")]
LocalShell {},
#[serde(rename = "custom")]
Freeform(FreeformTool),
}
#[derive(Debug, Clone)]
@@ -43,7 +62,7 @@ pub enum ConfigShellToolType {
pub struct ToolsConfig {
pub shell_type: ConfigShellToolType,
pub plan_tool: bool,
pub apply_patch_tool: bool,
pub apply_patch_tool_type: Option<ApplyPatchToolType>,
}
impl ToolsConfig {
@@ -65,10 +84,22 @@ impl ToolsConfig {
}
}
let apply_patch_tool_type = match model_family.apply_patch_tool_type {
Some(ApplyPatchToolType::Freeform) => Some(ApplyPatchToolType::Freeform),
Some(ApplyPatchToolType::Function) => Some(ApplyPatchToolType::Function),
None => {
if include_apply_patch_tool {
Some(ApplyPatchToolType::Freeform)
} else {
None
}
}
};
Self {
shell_type,
plan_tool: include_plan_tool,
apply_patch_tool: include_apply_patch_tool || model_family.uses_apply_patch_tool,
apply_patch_tool_type,
}
}
}
@@ -241,102 +272,12 @@ The shell tool is used to execute shell commands.
},
})
}
/// TODO(dylan): deprecate once we get rid of json tool
#[derive(Serialize, Deserialize)]
pub(crate) struct ApplyPatchToolArgs {
pub(crate) input: String,
}
/// Returns a JSON tool that can be used to edit files. Public for testing, please use `get_openai_tools`.
fn create_apply_patch_json_tool() -> OpenAiTool {
let mut properties = BTreeMap::new();
properties.insert(
"input".to_string(),
JsonSchema::String {
description: Some(r#"The entire contents of the apply_patch command"#.to_string()),
},
);
OpenAiTool::Function(ResponsesApiTool {
name: "apply_patch".to_string(),
description: r#"Use the `apply_patch` tool to edit files.
Your patch language is a strippeddown, fileoriented diff format designed to be easy to parse and safe to apply. You can think of it as a highlevel envelope:
*** Begin Patch
[ one or more file sections ]
*** End Patch
Within that envelope, you get a sequence of file operations.
You MUST include a header to specify the action you are taking.
Each operation starts with one of three headers:
*** Add File: <path> - create a new file. Every following line is a + line (the initial contents).
*** Delete File: <path> - remove an existing file. Nothing follows.
*** Update File: <path> - patch an existing file in place (optionally with a rename).
May be immediately followed by *** Move to: <new path> if you want to rename the file.
Then one or more “hunks”, each introduced by @@ (optionally followed by a hunk header).
Within a hunk each line starts with:
For instructions on [context_before] and [context_after]:
- By default, show 3 lines of code immediately above and 3 lines immediately below each change. If a change is within 3 lines of a previous change, do NOT duplicate the first changes [context_after] lines in the second changes [context_before] lines.
- If 3 lines of context is insufficient to uniquely identify the snippet of code within the file, use the @@ operator to indicate the class or function to which the snippet belongs. For instance, we might have:
@@ class BaseClass
[3 lines of pre-context]
- [old_code]
+ [new_code]
[3 lines of post-context]
- If a code block is repeated so many times in a class or function such that even a single `@@` statement and 3 lines of context cannot uniquely identify the snippet of code, you can use multiple `@@` statements to jump to the right context. For instance:
@@ class BaseClass
@@ def method():
[3 lines of pre-context]
- [old_code]
+ [new_code]
[3 lines of post-context]
The full grammar definition is below:
Patch := Begin { FileOp } End
Begin := "*** Begin Patch" NEWLINE
End := "*** End Patch" NEWLINE
FileOp := AddFile | DeleteFile | UpdateFile
AddFile := "*** Add File: " path NEWLINE { "+" line NEWLINE }
DeleteFile := "*** Delete File: " path NEWLINE
UpdateFile := "*** Update File: " path NEWLINE [ MoveTo ] { Hunk }
MoveTo := "*** Move to: " newPath NEWLINE
Hunk := "@@" [ header ] NEWLINE { HunkLine } [ "*** End of File" NEWLINE ]
HunkLine := (" " | "-" | "+") text NEWLINE
A full patch can combine several operations:
*** Begin Patch
*** Add File: hello.txt
+Hello world
*** Update File: src/app.py
*** Move to: src/main.py
@@ def greet():
-print("Hi")
+print("Hello, world!")
*** Delete File: obsolete.txt
*** End Patch
It is important to remember:
- You must include a header with your intended action (Add/Delete/Update)
- You must prefix new lines with `+` even when creating a new file
- File references can only be relative, NEVER ABSOLUTE.
"#
.to_string(),
strict: false,
parameters: JsonSchema::Object {
properties,
required: Some(vec!["input".to_string()]),
additional_properties: Some(false),
},
})
}
/// Returns JSON values that are compatible with Function Calling in the
/// Responses API:
/// https://platform.openai.com/docs/guides/function-calling?api-mode=responses
@@ -557,8 +498,15 @@ pub(crate) fn get_openai_tools(
tools.push(PLAN_TOOL.clone());
}
if config.apply_patch_tool {
tools.push(create_apply_patch_json_tool());
if let Some(apply_patch_tool_type) = &config.apply_patch_tool_type {
match apply_patch_tool_type {
ApplyPatchToolType::Freeform => {
tools.push(create_apply_patch_freeform_tool());
}
ApplyPatchToolType::Function => {
tools.push(create_apply_patch_json_tool());
}
}
}
if let Some(mcp_tools) = mcp_tools {
@@ -589,6 +537,7 @@ mod tests {
.map(|tool| match tool {
OpenAiTool::Function(ResponsesApiTool { name, .. }) => name,
OpenAiTool::LocalShell {} => "local_shell",
OpenAiTool::Freeform(FreeformTool { name, .. }) => name,
})
.collect::<Vec<_>>();
@@ -614,7 +563,7 @@ mod tests {
AskForApproval::Never,
SandboxPolicy::ReadOnly,
true,
model_family.uses_apply_patch_tool,
false,
);
let tools = get_openai_tools(&config, Some(HashMap::new()));
@@ -629,7 +578,7 @@ mod tests {
AskForApproval::Never,
SandboxPolicy::ReadOnly,
true,
model_family.uses_apply_patch_tool,
false,
);
let tools = get_openai_tools(&config, Some(HashMap::new()));
@@ -644,7 +593,7 @@ mod tests {
AskForApproval::Never,
SandboxPolicy::ReadOnly,
false,
model_family.uses_apply_patch_tool,
false,
);
let tools = get_openai_tools(
&config,
@@ -738,7 +687,7 @@ mod tests {
AskForApproval::Never,
SandboxPolicy::ReadOnly,
false,
model_family.uses_apply_patch_tool,
false,
);
let tools = get_openai_tools(
@@ -794,7 +743,7 @@ mod tests {
AskForApproval::Never,
SandboxPolicy::ReadOnly,
false,
model_family.uses_apply_patch_tool,
false,
);
let tools = get_openai_tools(
@@ -845,7 +794,7 @@ mod tests {
AskForApproval::Never,
SandboxPolicy::ReadOnly,
false,
model_family.uses_apply_patch_tool,
false,
);
let tools = get_openai_tools(
@@ -899,7 +848,7 @@ mod tests {
AskForApproval::Never,
SandboxPolicy::ReadOnly,
false,
model_family.uses_apply_patch_tool,
false,
);
let tools = get_openai_tools(

View File

@@ -132,6 +132,8 @@ impl RolloutRecorder {
| ResponseItem::LocalShellCall { .. }
| ResponseItem::FunctionCall { .. }
| ResponseItem::FunctionCallOutput { .. }
| ResponseItem::CustomToolCall { .. }
| ResponseItem::CustomToolCallOutput { .. }
| ResponseItem::Reasoning { .. } => filtered.push(item.clone()),
ResponseItem::Other => {
// These should never be serialized.
@@ -194,6 +196,8 @@ impl RolloutRecorder {
| ResponseItem::LocalShellCall { .. }
| ResponseItem::FunctionCall { .. }
| ResponseItem::FunctionCallOutput { .. }
| ResponseItem::CustomToolCall { .. }
| ResponseItem::CustomToolCallOutput { .. }
| ResponseItem::Reasoning { .. } => items.push(item),
ResponseItem::Other => {}
},
@@ -317,6 +321,8 @@ async fn rollout_writer(
| ResponseItem::LocalShellCall { .. }
| ResponseItem::FunctionCall { .. }
| ResponseItem::FunctionCallOutput { .. }
| ResponseItem::CustomToolCall { .. }
| ResponseItem::CustomToolCallOutput { .. }
| ResponseItem::Reasoning { .. } => {
writer.write_line(&item).await?;
}

View File

@@ -0,0 +1,145 @@
use serde::Deserialize;
use serde::Serialize;
use std::collections::BTreeMap;
use crate::openai_tools::FreeformTool;
use crate::openai_tools::FreeformToolFormat;
use crate::openai_tools::JsonSchema;
use crate::openai_tools::OpenAiTool;
use crate::openai_tools::ResponsesApiTool;
#[derive(Serialize, Deserialize)]
pub(crate) struct ApplyPatchToolArgs {
pub(crate) input: String,
}
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, Eq, Hash)]
#[serde(rename_all = "snake_case")]
pub enum ApplyPatchToolType {
Freeform,
Function,
}
/// Returns a custom tool that can be used to edit files. Well-suited for GPT-5 models
/// https://platform.openai.com/docs/guides/function-calling#custom-tools
pub(crate) fn create_apply_patch_freeform_tool() -> OpenAiTool {
OpenAiTool::Freeform(FreeformTool {
name: "apply_patch".to_string(),
description: "Use the `apply_patch` tool to edit files".to_string(),
format: FreeformToolFormat {
r#type: "grammar".to_string(),
syntax: "lark".to_string(),
definition: r#"start: begin_patch hunk+ end_patch
begin_patch: "*** Begin Patch" LF
end_patch: "*** End Patch" LF?
hunk: add_hunk | delete_hunk | update_hunk
add_hunk: "*** Add File: " filename LF add_line+
delete_hunk: "*** Delete File: " filename LF
update_hunk: "*** Update File: " filename LF change_move? change?
filename: /(.+)/
add_line: "+" /(.+)/ LF -> line
change_move: "*** Move to: " filename LF
change: (change_context | change_line)+ eof_line?
change_context: ("@@" | "@@ " /(.+)/) LF
change_line: ("+" | "-" | " ") /(.+)/ LF
eof_line: "*** End of File" LF
%import common.LF
"#
.to_string(),
},
})
}
/// Returns a json tool that can be used to edit files. Should only be used with gpt-oss models
pub(crate) fn create_apply_patch_json_tool() -> OpenAiTool {
let mut properties = BTreeMap::new();
properties.insert(
"input".to_string(),
JsonSchema::String {
description: Some(r#"The entire contents of the apply_patch command"#.to_string()),
},
);
OpenAiTool::Function(ResponsesApiTool {
name: "apply_patch".to_string(),
description: r#"Use the `apply_patch` tool to edit files.
Your patch language is a strippeddown, fileoriented diff format designed to be easy to parse and safe to apply. You can think of it as a highlevel envelope:
*** Begin Patch
[ one or more file sections ]
*** End Patch
Within that envelope, you get a sequence of file operations.
You MUST include a header to specify the action you are taking.
Each operation starts with one of three headers:
*** Add File: <path> - create a new file. Every following line is a + line (the initial contents).
*** Delete File: <path> - remove an existing file. Nothing follows.
*** Update File: <path> - patch an existing file in place (optionally with a rename).
May be immediately followed by *** Move to: <new path> if you want to rename the file.
Then one or more “hunks”, each introduced by @@ (optionally followed by a hunk header).
Within a hunk each line starts with:
For instructions on [context_before] and [context_after]:
- By default, show 3 lines of code immediately above and 3 lines immediately below each change. If a change is within 3 lines of a previous change, do NOT duplicate the first changes [context_after] lines in the second changes [context_before] lines.
- If 3 lines of context is insufficient to uniquely identify the snippet of code within the file, use the @@ operator to indicate the class or function to which the snippet belongs. For instance, we might have:
@@ class BaseClass
[3 lines of pre-context]
- [old_code]
+ [new_code]
[3 lines of post-context]
- If a code block is repeated so many times in a class or function such that even a single `@@` statement and 3 lines of context cannot uniquely identify the snippet of code, you can use multiple `@@` statements to jump to the right context. For instance:
@@ class BaseClass
@@ def method():
[3 lines of pre-context]
- [old_code]
+ [new_code]
[3 lines of post-context]
The full grammar definition is below:
Patch := Begin { FileOp } End
Begin := "*** Begin Patch" NEWLINE
End := "*** End Patch" NEWLINE
FileOp := AddFile | DeleteFile | UpdateFile
AddFile := "*** Add File: " path NEWLINE { "+" line NEWLINE }
DeleteFile := "*** Delete File: " path NEWLINE
UpdateFile := "*** Update File: " path NEWLINE [ MoveTo ] { Hunk }
MoveTo := "*** Move to: " newPath NEWLINE
Hunk := "@@" [ header ] NEWLINE { HunkLine } [ "*** End of File" NEWLINE ]
HunkLine := (" " | "-" | "+") text NEWLINE
A full patch can combine several operations:
*** Begin Patch
*** Add File: hello.txt
+Hello world
*** Update File: src/app.py
*** Move to: src/main.py
@@ def greet():
-print("Hi")
+print("Hello, world!")
*** Delete File: obsolete.txt
*** End Patch
It is important to remember:
- You must include a header with your intended action (Add/Delete/Update)
- You must prefix new lines with `+` even when creating a new file
- File references can only be relative, NEVER ABSOLUTE.
"#
.to_string(),
strict: false,
parameters: JsonSchema::Object {
properties,
required: Some(vec!["input".to_string()]),
additional_properties: Some(false),
},
})
}

View File

@@ -1,6 +1,9 @@
#![allow(clippy::unwrap_used)]
use codex_core::ConversationManager;
use codex_core::ModelProviderInfo;
use codex_core::built_in_model_providers;
use codex_core::model_family::find_family_for_model;
use codex_core::protocol::AskForApproval;
use codex_core::protocol::EventMsg;
use codex_core::protocol::InputItem;
@@ -25,8 +28,20 @@ fn sse_completed(id: &str) -> String {
load_sse_fixture_with_id("tests/fixtures/completed_template.json", id)
}
fn assert_tool_names(body: &serde_json::Value, expected_names: &[&str]) {
assert_eq!(
body["tools"]
.as_array()
.unwrap()
.iter()
.map(|t| t["name"].as_str().unwrap().to_string())
.collect::<Vec<_>>(),
expected_names
);
}
#[tokio::test(flavor = "multi_thread", worker_threads = 4)]
async fn default_system_instructions_contain_apply_patch() {
async fn codex_mini_latest_tools() {
use pretty_assertions::assert_eq;
let server = MockServer::start().await;
@@ -58,6 +73,10 @@ async fn default_system_instructions_contain_apply_patch() {
let conversation_manager =
ConversationManager::with_auth(CodexAuth::from_api_key("Test API Key"));
config.include_apply_patch_tool = false;
config.model = "codex-mini-latest".to_string();
config.model_family = find_family_for_model("codex-mini-latest").unwrap();
let codex = conversation_manager
.new_conversation(config)
.await
@@ -173,18 +192,6 @@ async fn prompt_tools_are_consistent_across_requests() {
// our internal implementation is responsible for keeping tools in sync
// with the OpenAI schema, so we just verify the tool presence here
let expected_tools_names: &[&str] = &["shell", "update_plan", "apply_patch"];
fn assert_tool_names(body: &serde_json::Value, expected_names: &[&str]) {
assert_eq!(
body["tools"]
.as_array()
.unwrap()
.iter()
.map(|t| t["name"].as_str().unwrap().to_string())
.collect::<Vec<_>>(),
expected_names
);
}
let body0 = requests[0].body_json::<serde_json::Value>().unwrap();
assert_eq!(
body0["instructions"],

View File

@@ -123,6 +123,155 @@ async fn test_apply_patch_tool() -> anyhow::Result<()> {
// Start a mock model server
let server = MockServer::start().await;
// First response: model calls apply_patch to create test.md
let first = ResponseTemplate::new(200)
.insert_header("content-type", "text/event-stream")
.set_body_raw(
load_sse_fixture_with_id_from_str(SSE_TOOL_CALL_ADD, "call1"),
"text/event-stream",
);
Mock::given(method("POST"))
// .and(path("/v1/responses"))
.respond_with(first)
.up_to_n_times(1)
.mount(&server)
.await;
// Second response: model calls apply_patch to update test.md
let second = ResponseTemplate::new(200)
.insert_header("content-type", "text/event-stream")
.set_body_raw(
load_sse_fixture_with_id_from_str(SSE_TOOL_CALL_UPDATE, "call2"),
"text/event-stream",
);
Mock::given(method("POST"))
.and(path("/v1/responses"))
.respond_with(second)
.up_to_n_times(1)
.mount(&server)
.await;
let final_completed = ResponseTemplate::new(200)
.insert_header("content-type", "text/event-stream")
.set_body_raw(
load_sse_fixture_with_id_from_str(SSE_TOOL_CALL_COMPLETED, "resp3"),
"text/event-stream",
);
Mock::given(method("POST"))
.and(path("/v1/responses"))
.respond_with(final_completed)
.expect(1)
.mount(&server)
.await;
let tmp_cwd = TempDir::new().unwrap();
Command::cargo_bin("codex-exec")
.context("should find binary for codex-exec")?
.current_dir(tmp_cwd.path())
.env("CODEX_HOME", tmp_cwd.path())
.env("OPENAI_API_KEY", "dummy")
.env("OPENAI_BASE_URL", format!("{}/v1", server.uri()))
.arg("--skip-git-repo-check")
.arg("-s")
.arg("workspace-write")
.arg("foo")
.assert()
.success();
// Verify final file contents
let final_path = tmp_cwd.path().join("test.md");
let contents = std::fs::read_to_string(&final_path)
.unwrap_or_else(|e| panic!("failed reading {}: {e}", final_path.display()));
assert_eq!(contents, "Final text\n");
Ok(())
}
#[cfg(not(target_os = "windows"))]
#[tokio::test]
async fn test_apply_patch_freeform_tool() -> anyhow::Result<()> {
use core_test_support::load_sse_fixture_with_id_from_str;
use tempfile::TempDir;
use wiremock::Mock;
use wiremock::MockServer;
use wiremock::ResponseTemplate;
use wiremock::matchers::method;
use wiremock::matchers::path;
const SSE_TOOL_CALL_ADD: &str = r#"[
{
"type": "response.output_item.done",
"item": {
"type": "custom_tool_call",
"name": "apply_patch",
"input": "*** Begin Patch\n*** Add File: test.md\n+Hello world\n*** End Patch",
"call_id": "__ID__"
}
},
{
"type": "response.completed",
"response": {
"id": "__ID__",
"usage": {
"input_tokens": 0,
"input_tokens_details": null,
"output_tokens": 0,
"output_tokens_details": null,
"total_tokens": 0
},
"output": []
}
}
]"#;
const SSE_TOOL_CALL_UPDATE: &str = r#"[
{
"type": "response.output_item.done",
"item": {
"type": "custom_tool_call",
"name": "apply_patch",
"input": "*** Begin Patch\n*** Update File: test.md\n@@\n-Hello world\n+Final text\n*** End Patch",
"call_id": "__ID__"
}
},
{
"type": "response.completed",
"response": {
"id": "__ID__",
"usage": {
"input_tokens": 0,
"input_tokens_details": null,
"output_tokens": 0,
"output_tokens_details": null,
"total_tokens": 0
},
"output": []
}
}
]"#;
const SSE_TOOL_CALL_COMPLETED: &str = r#"[
{
"type": "response.completed",
"response": {
"id": "__ID__",
"usage": {
"input_tokens": 0,
"input_tokens_details": null,
"output_tokens": 0,
"output_tokens_details": null,
"total_tokens": 0
},
"output": []
}
}
]"#;
// Start a mock model server
let server = MockServer::start().await;
// First response: model calls apply_patch to create test.md
let first = ResponseTemplate::new(200)
.insert_header("content-type", "text/event-stream")