PicoBot/src/agent/agent_loop.rs
2026-04-22 06:57:22 +08:00

770 lines
27 KiB
Rust

use crate::bus::message::ContentBlock;
use crate::bus::ChatMessage;
use crate::config::LLMProviderConfig;
use crate::observability::{
truncate_args, Observer, ObserverEvent, ToolExecutionOutcome,
};
use crate::providers::{create_provider, LLMProvider, ChatCompletionRequest, Message, ToolCall};
use crate::skills::SkillRuntime;
use crate::storage::SessionStore;
use crate::tools::{ToolContext, ToolRegistry};
use std::collections::VecDeque;
use std::hash::{Hash, Hasher};
use std::io::Read;
use std::sync::Arc;
use std::time::Instant;
/// Maximum characters in a tool result before truncation.
/// Prevents context overflow from large tool outputs.
const MAX_TOOL_RESULT_CHARS: usize = 16_000;
/// Minimum characters to keep when truncating
const TRUNCATION_SUFFIX_LEN: usize = 200;
/// Build content blocks from text and media paths
fn build_content_blocks(text: &str, media_paths: &[String]) -> Vec<ContentBlock> {
let mut blocks = Vec::new();
// Add text block if there's text
if !text.is_empty() {
blocks.push(ContentBlock::text(text));
}
// Add image blocks for media paths
for path in media_paths {
if let Ok((mime_type, base64_data)) = encode_image_to_base64(path) {
let url = format!("data:{};base64,{}", mime_type, base64_data);
blocks.push(ContentBlock::image_url(url));
}
}
// If nothing, add empty text block
if blocks.is_empty() {
blocks.push(ContentBlock::text(""));
}
blocks
}
/// Encode an image file to base64 data URL
fn encode_image_to_base64(path: &str) -> Result<(String, String), std::io::Error> {
use base64::{Engine as _, engine::general_purpose::STANDARD};
let mut file = std::fs::File::open(path)?;
let mut buffer = Vec::new();
file.read_to_end(&mut buffer)?;
let mime = mime_guess::from_path(path)
.first_or_octet_stream()
.to_string();
let encoded = STANDARD.encode(&buffer);
Ok((mime, encoded))
}
/// Truncate tool result if it exceeds MAX_TOOL_RESULT_CHARS.
/// Preserves the end of the output as it often contains the conclusion/useful result.
fn truncate_tool_result(output: &str) -> String {
if output.len() <= MAX_TOOL_RESULT_CHARS {
return output.to_string();
}
let truncated_start_len = output.len().saturating_sub(TRUNCATION_SUFFIX_LEN);
if truncated_start_len > MAX_TOOL_RESULT_CHARS {
// Even after removing suffix, still too long - take from beginning
format!(
"{}...\n\n[Output truncated - {} characters removed]",
&output[..MAX_TOOL_RESULT_CHARS - 100],
output.len() - MAX_TOOL_RESULT_CHARS + 100
)
} else {
// Keep most of the end which usually contains the useful result
format!(
"...\n\n[Output truncated - {} characters removed]\n\n{}",
truncated_start_len,
&output[truncated_start_len..]
)
}
}
/// Loop detection result.
#[derive(Debug, Clone, PartialEq, Eq)]
enum LoopDetectionResult {
/// No warning needed.
Ok,
/// Warning: same tool + args repeated N times.
Warning(String),
}
/// Configuration for loop detector.
#[derive(Debug, Clone)]
struct LoopDetectorConfig {
/// Master switch.
enabled: bool,
/// Warn every N consecutive identical calls.
warn_every: usize,
}
impl Default for LoopDetectorConfig {
fn default() -> Self {
Self {
enabled: true,
warn_every: 5,
}
}
}
/// A single recorded tool invocation in the sliding window.
#[derive(Debug, Clone)]
struct ToolCallRecord {
name: String,
args_hash: u64,
}
/// Stateful loop detector that monitors for repetitive patterns.
struct LoopDetector {
config: LoopDetectorConfig,
window: VecDeque<ToolCallRecord>,
}
impl LoopDetector {
fn new(config: LoopDetectorConfig) -> Self {
Self {
window: VecDeque::with_capacity(config.warn_every * 2),
config,
}
}
/// Record a completed tool call and check for loop patterns.
/// Returns Warning every `warn_every` consecutive identical calls.
fn record(&mut self, name: &str, args: &serde_json::Value) -> LoopDetectionResult {
if !self.config.enabled {
return LoopDetectionResult::Ok;
}
let record = ToolCallRecord {
name: name.to_string(),
args_hash: hash_json_value(args),
};
// Maintain sliding window
if self.window.len() >= self.config.warn_every * 2 {
self.window.pop_front();
}
self.window.push_back(record);
// Count consecutive identical calls
let last = self.window.back().unwrap();
let consecutive: usize = self
.window
.iter()
.rev()
.take_while(|r| r.name == last.name && r.args_hash == last.args_hash)
.count();
// Warn every warn_every times
if consecutive > 0 && consecutive % self.config.warn_every == 0 {
LoopDetectionResult::Warning(format!(
"注意: 工具 '{}' 已连续执行 {} 次,参数相同。如果任务没有进展,请尝试其他方法。",
last.name, consecutive
))
} else {
LoopDetectionResult::Ok
}
}
}
/// Hash a JSON value deterministically (key-order independent).
fn hash_json_value(value: &serde_json::Value) -> u64 {
let mut hasher = std::collections::hash_map::DefaultHasher::new();
let canonical = canonicalise_json(value);
canonical.hash(&mut hasher);
hasher.finish()
}
/// Return a clone of value with all object keys sorted recursively.
fn canonicalise_json(value: &serde_json::Value) -> serde_json::Value {
match value {
serde_json::Value::Object(map) => {
let mut sorted: Vec<(&String, &serde_json::Value)> = map.iter().collect();
sorted.sort_by_key(|(k, _)| *k);
let new_map: serde_json::Map<String, serde_json::Value> = sorted
.into_iter()
.map(|(k, v)| (k.clone(), canonicalise_json(v)))
.collect();
serde_json::Value::Object(new_map)
}
serde_json::Value::Array(arr) => {
serde_json::Value::Array(arr.iter().map(canonicalise_json).collect())
}
other => other.clone(),
}
}
/// Convert ChatMessage to LLM Message format
fn chat_message_to_llm_message(m: &ChatMessage) -> Message {
let content = if m.media_refs.is_empty() {
vec![ContentBlock::text(&m.content)]
} else {
build_content_blocks(&m.content, &m.media_refs)
};
Message {
role: m.role.clone(),
content,
tool_call_id: m.tool_call_id.clone(),
name: m.tool_name.clone(),
tool_calls: m.tool_calls.clone(),
}
}
/// AgentLoop - Stateless agent that processes messages with tool calling support.
/// History is managed externally by SessionManager.
pub struct AgentLoop {
provider: Box<dyn LLMProvider>,
tools: Arc<ToolRegistry>,
skills: Arc<SkillRuntime>,
skill_event_store: Option<Arc<SessionStore>>,
skill_event_session_id: Option<String>,
tool_context: ToolContext,
observer: Option<Arc<dyn Observer>>,
max_iterations: usize,
}
#[derive(Debug, Clone)]
pub struct AgentProcessResult {
pub final_response: ChatMessage,
pub emitted_messages: Vec<ChatMessage>,
}
impl AgentLoop {
pub fn new(provider_config: LLMProviderConfig) -> Result<Self, AgentError> {
let max_iterations = provider_config.max_tool_iterations;
let provider = create_provider(provider_config)
.map_err(|e| AgentError::ProviderCreation(e.to_string()))?;
Ok(Self {
provider,
tools: Arc::new(ToolRegistry::new()),
skills: Arc::new(SkillRuntime::default()),
skill_event_store: None,
skill_event_session_id: None,
tool_context: ToolContext::default(),
observer: None,
max_iterations,
})
}
pub fn with_tools(provider_config: LLMProviderConfig, tools: Arc<ToolRegistry>) -> Result<Self, AgentError> {
let max_iterations = provider_config.max_tool_iterations;
let provider = create_provider(provider_config)
.map_err(|e| AgentError::ProviderCreation(e.to_string()))?;
Ok(Self {
provider,
tools,
skills: Arc::new(SkillRuntime::default()),
skill_event_store: None,
skill_event_session_id: None,
tool_context: ToolContext::default(),
observer: None,
max_iterations,
})
}
pub fn with_tools_and_skills(
provider_config: LLMProviderConfig,
tools: Arc<ToolRegistry>,
skills: Arc<SkillRuntime>,
) -> Result<Self, AgentError> {
let max_iterations = provider_config.max_tool_iterations;
let provider = create_provider(provider_config)
.map_err(|e| AgentError::ProviderCreation(e.to_string()))?;
Ok(Self {
provider,
tools,
skills,
skill_event_store: None,
skill_event_session_id: None,
tool_context: ToolContext::default(),
observer: None,
max_iterations,
})
}
pub fn with_skill_event_store(mut self, store: Arc<SessionStore>, session_id: String) -> Self {
self.skill_event_store = Some(store);
self.skill_event_session_id = Some(session_id);
self
}
pub fn with_tool_context(mut self, context: ToolContext) -> Self {
self.tool_context = context;
self
}
/// Set an observer for tracking events.
pub fn with_observer(mut self, observer: Arc<dyn Observer>) -> Self {
self.observer = Some(observer);
self
}
pub fn tools(&self) -> &Arc<ToolRegistry> {
&self.tools
}
/// Process a message using the provided conversation history.
/// History management is handled externally by SessionManager.
///
/// This method supports multi-round tool calling: after executing tools,
/// it loops back to the LLM with the tool results until either:
/// - The LLM returns no more tool calls (final response)
/// - Maximum iterations are reached
pub async fn process(&self, mut messages: Vec<ChatMessage>) -> Result<AgentProcessResult, AgentError> {
#[cfg(debug_assertions)]
tracing::debug!(history_len = messages.len(), max_iterations = self.max_iterations, "Starting agent process");
// Track tool calls for loop detection
let mut loop_detector = LoopDetector::new(LoopDetectorConfig::default());
let mut emitted_messages = Vec::new();
for iteration in 0..self.max_iterations {
#[cfg(debug_assertions)]
tracing::debug!(iteration, "Agent iteration started");
// Convert messages to LLM format
let mut messages_for_llm: Vec<Message> = Vec::with_capacity(messages.len() + 1);
if let Some(skill_prompt) = self.skills.system_index_prompt() {
messages_for_llm.push(Message::system(skill_prompt));
}
messages_for_llm.extend(messages.iter().map(chat_message_to_llm_message));
// Build request
let mut tool_defs = self.tools.get_definitions();
if let Some(skill_tool) = self.skills.skill_tool_definition() {
tool_defs.push(skill_tool);
}
let tools = if tool_defs.is_empty() { None } else { Some(tool_defs) };
let request = ChatCompletionRequest {
messages: messages_for_llm,
temperature: None,
max_tokens: None,
tools,
};
// Call LLM
let response = (*self.provider).chat(request).await
.map_err(|e| {
tracing::error!(error = %e, "LLM request failed");
AgentError::LlmError(e.to_string())
})?;
#[cfg(debug_assertions)]
tracing::debug!(
iteration,
response_len = response.content.len(),
tool_calls_len = response.tool_calls.len(),
"LLM response received"
);
// If no tool calls, this is the final response
if response.tool_calls.is_empty() {
let assistant_message = ChatMessage::assistant(response.content);
emitted_messages.push(assistant_message.clone());
return Ok(AgentProcessResult {
final_response: assistant_message,
emitted_messages,
});
}
// Execute tool calls
tracing::info!(iteration, count = response.tool_calls.len(), "Tool calls detected, executing tools");
// Add assistant message with tool calls
let assistant_message = ChatMessage::assistant_with_tool_calls(
response.content.clone(),
response.tool_calls.clone(),
);
messages.push(assistant_message.clone());
emitted_messages.push(assistant_message);
// Execute tools and add results to messages
let tool_results = self.execute_tools(&response.tool_calls).await;
for (tool_call, result) in response.tool_calls.iter().zip(tool_results.iter()) {
// Log function call with name and arguments
let args_str = match &tool_call.arguments {
serde_json::Value::Object(obj) if obj.is_empty() => "{}".to_string(),
other => serde_json::to_string_pretty(other).unwrap_or_else(|_| other.to_string()),
};
tracing::info!(tool = %tool_call.name, args = %args_str, "Calling tool");
// Truncate tool result if too large
let truncated_output = truncate_tool_result(&result.output);
// Record tool call and check for loops
let loop_result = loop_detector.record(&tool_call.name, &tool_call.arguments);
match loop_result {
LoopDetectionResult::Warning(msg) => {
// Add warning and proceed
tracing::warn!(
tool = %tool_call.name,
"Loop warning: {}",
msg
);
let tool_message = ChatMessage::tool(
tool_call.id.clone(),
tool_call.name.clone(),
format!("{}\n\n[上一条结果]\n{}", msg, truncated_output),
);
messages.push(tool_message.clone());
emitted_messages.push(tool_message);
}
LoopDetectionResult::Ok => {
let tool_message = ChatMessage::tool(
tool_call.id.clone(),
tool_call.name.clone(),
truncated_output,
);
messages.push(tool_message.clone());
emitted_messages.push(tool_message);
}
}
}
// Loop continues to next iteration with updated messages
#[cfg(debug_assertions)]
tracing::debug!(iteration, message_count = messages.len(), "Tool execution complete, continuing to next iteration");
}
// Max iterations reached - ask LLM for a summary based on completed work
tracing::warn!("Max iterations reached, requesting final summary from LLM");
// Add a message asking for summary
let summary_request = ChatMessage::user(
"You have reached the maximum number of tool call iterations. \
Please provide your best answer based on the work completed so far."
);
messages.push(summary_request);
// Convert messages to LLM format
let mut messages_for_llm: Vec<Message> = Vec::with_capacity(messages.len() + 1);
if let Some(skill_prompt) = self.skills.system_index_prompt() {
messages_for_llm.push(Message::system(skill_prompt));
}
messages_for_llm.extend(messages.iter().map(chat_message_to_llm_message));
let request = ChatCompletionRequest {
messages: messages_for_llm,
temperature: None,
max_tokens: None,
tools: None, // No tools in final summary call
};
match (*self.provider).chat(request).await {
Ok(response) => {
let assistant_message = ChatMessage::assistant(response.content);
emitted_messages.push(assistant_message.clone());
Ok(AgentProcessResult {
final_response: assistant_message,
emitted_messages,
})
}
Err(e) => {
// Fallback if summary call fails
tracing::error!(error = %e, "Failed to get summary from LLM");
let final_message = ChatMessage::assistant(
format!("I reached the maximum number of tool call iterations ({}) without completing the task. The work done so far has been lost due to an error. Please try breaking the task into smaller steps.", self.max_iterations)
);
emitted_messages.push(final_message.clone());
Ok(AgentProcessResult {
final_response: final_message,
emitted_messages,
})
}
}
}
/// Determine whether to execute tools in parallel or sequentially.
///
/// Returns true if:
/// - There are multiple tool calls
/// - None of the tools require sequential execution (tool_search, non-concurrency-safe)
fn should_execute_in_parallel(&self, tool_calls: &[ToolCall]) -> bool {
if tool_calls.len() <= 1 {
return false;
}
// tool_search must run sequentially to avoid MCP activation race conditions
if tool_calls.iter().any(|tc| tc.name == "tool_search") {
return false;
}
// All tools must be concurrency-safe to run in parallel
tool_calls.iter().all(|tc| {
self.tools
.get(&tc.name)
.map(|t| t.concurrency_safe())
.unwrap_or(false)
})
}
/// Execute multiple tool calls, choosing parallel or sequential based on conditions.
async fn execute_tools(&self, tool_calls: &[ToolCall]) -> Vec<ToolExecutionOutcome> {
if self.should_execute_in_parallel(tool_calls) {
tracing::debug!("Executing {} tools in parallel", tool_calls.len());
self.execute_tools_parallel(tool_calls).await
} else {
tracing::debug!("Executing {} tools sequentially", tool_calls.len());
self.execute_tools_sequential(tool_calls).await
}
}
/// Execute tools in parallel using join_all.
async fn execute_tools_parallel(&self, tool_calls: &[ToolCall]) -> Vec<ToolExecutionOutcome> {
let futures: Vec<_> = tool_calls
.iter()
.map(|tc| self.execute_one_tool(tc))
.collect();
futures_util::future::join_all(futures).await
}
/// Execute tools sequentially.
async fn execute_tools_sequential(&self, tool_calls: &[ToolCall]) -> Vec<ToolExecutionOutcome> {
let mut outcomes = Vec::with_capacity(tool_calls.len());
for tool_call in tool_calls {
outcomes.push(self.execute_one_tool(tool_call).await);
}
outcomes
}
/// Execute a single tool and return the outcome with event tracking.
async fn execute_one_tool(&self, tool_call: &ToolCall) -> ToolExecutionOutcome {
let start = Instant::now();
let tool_name = tool_call.name.clone();
// Record ToolCallStart event
if let Some(ref observer) = self.observer {
observer.record_event(&ObserverEvent::ToolCallStart {
tool: tool_name.clone(),
arguments: Some(truncate_args(&tool_call.arguments, 300)),
});
}
let result = self.execute_tool_internal(tool_call).await;
let duration = start.elapsed();
// Record ToolCall event
if let Some(ref observer) = self.observer {
observer.record_event(&ObserverEvent::ToolCall {
tool: tool_name.clone(),
duration,
success: result.success,
});
}
// Apply duration
ToolExecutionOutcome {
duration,
..result
}
}
/// Internal tool execution without event tracking.
async fn execute_tool_internal(&self, tool_call: &ToolCall) -> ToolExecutionOutcome {
if tool_call.name == "skill_activate" {
let skill_name = match tool_call.arguments.get("name").and_then(|v| v.as_str()) {
Some(name) if !name.trim().is_empty() => name,
_ => {
self.record_skill_event(
"activation_failed",
None,
serde_json::json!({
"reason": "missing_name",
"arguments": tool_call.arguments,
}),
);
return ToolExecutionOutcome::failure(
"Error: Missing required parameter: name".to_string(),
Some("Missing required parameter: name".to_string()),
);
}
};
return match self.skills.activation_payload(skill_name) {
Ok(output) => {
if let Ok(payload) = self.skills.activation_event_payload(skill_name) {
self.record_skill_event("activated", Some(skill_name), payload);
}
ToolExecutionOutcome::success(output)
}
Err(err) => {
self.record_skill_event(
"activation_failed",
Some(skill_name),
serde_json::json!({
"reason": err,
"arguments": tool_call.arguments,
}),
);
ToolExecutionOutcome::failure(
format!("Error: {}", err),
Some(err),
)
}
};
}
let tool = match self.tools.get(&tool_call.name) {
Some(t) => t,
None => {
tracing::warn!(tool = %tool_call.name, "Tool not found");
return ToolExecutionOutcome::failure(
format!("Error: Tool '{}' not found", tool_call.name),
Some(format!("Tool '{}' not found", tool_call.name)),
);
}
};
match tool.execute_with_context(&self.tool_context, tool_call.arguments.clone()).await {
Ok(result) => {
if result.success {
ToolExecutionOutcome::success(result.output)
} else {
let error = result.error.unwrap_or_default();
ToolExecutionOutcome::failure(
format!("Error: {}", error),
Some(error),
)
}
}
Err(e) => {
tracing::error!(tool = %tool_call.name, error = %e, "Tool execution failed");
ToolExecutionOutcome::failure(
format!("Error: {}", e),
Some(e.to_string()),
)
}
}
}
fn record_skill_event(
&self,
event_type: &str,
skill_name: Option<&str>,
payload: serde_json::Value,
) {
let (Some(store), Some(session_id)) = (
self.skill_event_store.as_ref(),
self.skill_event_session_id.as_ref(),
) else {
return;
};
if let Err(err) = store.append_skill_event(Some(session_id), event_type, skill_name, &payload) {
tracing::warn!(error = %err, event_type = %event_type, "Failed to record skill event");
}
}
}
#[cfg(test)]
mod tests {
use super::*;
use crate::observability::{MultiObserver, Observer};
struct TestObserver {
events: std::sync::Mutex<Vec<ObserverEvent>>,
}
impl TestObserver {
fn new() -> Self {
Self {
events: std::sync::Mutex::new(Vec::new()),
}
}
}
impl Observer for TestObserver {
fn record_event(&self, event: &ObserverEvent) {
self.events.lock().unwrap().push(event.clone());
}
fn name(&self) -> &str {
"test_observer"
}
}
#[tokio::test]
async fn test_observer_receives_tool_events() {
// Verify MultiObserver works
let mut multi = MultiObserver::new();
multi.add_observer(Box::new(TestObserver::new()));
let event = ObserverEvent::ToolCallStart {
tool: "test".to_string(),
arguments: Some("{}".to_string()),
};
multi.record_event(&event);
// Just verify the structure works
assert_eq!(multi.len(), 1);
}
#[test]
fn test_should_execute_in_parallel_single_tool() {
// Would need a proper setup with AgentLoop to test fully
// For now, just verify the logic: single tool should return false
let calls = vec![ToolCall {
id: "1".to_string(),
name: "test".to_string(),
arguments: serde_json::json!({}),
}];
// If there's only 1 tool, should return false regardless
assert_eq!(calls.len() <= 1, true);
}
#[test]
fn test_chat_message_to_llm_message_preserves_assistant_tool_calls() {
let chat_message = ChatMessage::assistant_with_tool_calls(
"calling tool",
vec![ToolCall {
id: "call_1".to_string(),
name: "calculator".to_string(),
arguments: serde_json::json!({ "expression": "2+2" }),
}],
);
let provider_message = chat_message_to_llm_message(&chat_message);
assert_eq!(provider_message.role, "assistant");
assert_eq!(provider_message.tool_calls.as_ref().unwrap().len(), 1);
assert_eq!(provider_message.tool_calls.as_ref().unwrap()[0].id, "call_1");
assert_eq!(provider_message.tool_calls.as_ref().unwrap()[0].name, "calculator");
}
}
#[derive(Debug)]
pub enum AgentError {
ProviderCreation(String),
LlmError(String),
Other(String),
}
impl std::fmt::Display for AgentError {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
match self {
AgentError::ProviderCreation(e) => write!(f, "Provider creation error: {}", e),
AgentError::LlmError(e) => write!(f, "LLM error: {}", e),
AgentError::Other(e) => write!(f, "{}", e),
}
}
}
impl std::error::Error for AgentError {}