swiftide_agents/
agent.rs

1#![allow(dead_code)]
2use crate::{
3    default_context::DefaultContext,
4    hooks::{
5        AfterCompletionFn, AfterEachFn, AfterToolFn, BeforeAllFn, BeforeCompletionFn, BeforeToolFn,
6        Hook, HookTypes, MessageHookFn, OnStartFn,
7    },
8    state,
9    system_prompt::SystemPrompt,
10    tools::{arg_preprocessor::ArgPreprocessor, control::Stop},
11};
12use std::{
13    collections::{HashMap, HashSet},
14    hash::{DefaultHasher, Hash as _, Hasher as _},
15    sync::Arc,
16};
17
18use anyhow::Result;
19use derive_builder::Builder;
20use swiftide_core::{
21    chat_completion::{
22        ChatCompletion, ChatCompletionRequest, ChatMessage, Tool, ToolCall, ToolOutput,
23    },
24    prompt::Prompt,
25    AgentContext,
26};
27use tracing::{debug, Instrument};
28
29/// Agents are the main interface for building agentic systems.
30///
31/// Construct agents by calling the builder, setting an llm, configure hooks, tools and other
32/// customizations.
33///
34/// # Important defaults
35///
36/// - The default context is the `DefaultContext`, executing tools locally with the `LocalExecutor`.
37/// - A default `stop` tool is provided for agents to explicitly stop if needed
38/// - The default `SystemPrompt` instructs the agent with chain of thought and some common
39///   safeguards, but is otherwise quite bare. In a lot of cases this can be sufficient.
40#[derive(Clone, Builder)]
41pub struct Agent {
42    /// Hooks are functions that are called at specific points in the agent's lifecycle.
43    #[builder(default, setter(into))]
44    pub(crate) hooks: Vec<Hook>,
45    /// The context in which the agent operates, by default this is the `DefaultContext`.
46    #[builder(
47        setter(custom),
48        default = Arc::new(DefaultContext::default()) as Arc<dyn AgentContext>
49    )]
50    pub(crate) context: Arc<dyn AgentContext>,
51    /// Tools the agent can use
52    #[builder(default = Agent::default_tools(), setter(custom))]
53    pub(crate) tools: HashSet<Box<dyn Tool>>,
54
55    /// The language model that the agent uses for completion.
56    #[builder(setter(custom))]
57    pub(crate) llm: Box<dyn ChatCompletion>,
58
59    /// System prompt for the agent when it starts
60    ///
61    /// Some agents profit significantly from a tailored prompt. But it is not always needed.
62    ///
63    /// See [`SystemPrompt`] for an opiniated, customizable system prompt.
64    ///
65    /// Swiftide provides a default system prompt for all agents.
66    ///
67    /// # Example
68    ///
69    /// ```no_run
70    /// # use swiftide_agents::system_prompt::SystemPrompt;
71    /// # use swiftide_agents::Agent;
72    /// Agent::builder()
73    ///     .system_prompt(
74    ///         SystemPrompt::builder().role("You are an expert engineer")
75    ///         .build().unwrap())
76    ///     .build().unwrap();
77    /// ```
78    #[builder(setter(into, strip_option), default = Some(SystemPrompt::default().into()))]
79    pub(crate) system_prompt: Option<Prompt>,
80
81    /// Initial state of the agent
82    #[builder(private, default = state::State::default())]
83    pub(crate) state: state::State,
84
85    /// Optional limit on the amount of loops the agent can run.
86    /// The counter is reset when the agent is stopped.
87    #[builder(default, setter(strip_option))]
88    pub(crate) limit: Option<usize>,
89
90    /// The maximum amount of times the failed output of a tool will be send
91    /// to an LLM before the agent stops. Defaults to 3.
92    ///
93    /// LLMs sometimes send missing arguments, or a tool might actually fail, but retrying could be
94    /// worth while. If the limit is not reached, the agent will send the formatted error back to
95    /// the LLM.
96    ///
97    /// The limit is hashed based on the tool call name and arguments, so the limit is per tool
98    /// call.
99    ///
100    /// This limit is _not_ reset when the agent is stopped.
101    #[builder(default = 3)]
102    pub(crate) tool_retry_limit: usize,
103
104    /// Internally tracks the amount of times a tool has been retried. The key is a hash based on
105    /// the name and args of the tool.
106    #[builder(private, default)]
107    pub(crate) tool_retries_counter: HashMap<u64, usize>,
108}
109
110impl std::fmt::Debug for Agent {
111    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
112        f.debug_struct("Agent")
113            // display hooks as a list of type: number of hooks
114            .field(
115                "hooks",
116                &self
117                    .hooks
118                    .iter()
119                    .map(std::string::ToString::to_string)
120                    .collect::<Vec<_>>(),
121            )
122            .field(
123                "tools",
124                &self
125                    .tools
126                    .iter()
127                    .map(swiftide_core::Tool::name)
128                    .collect::<Vec<_>>(),
129            )
130            .field("llm", &"Box<dyn ChatCompletion>")
131            .field("state", &self.state)
132            .finish()
133    }
134}
135
136impl AgentBuilder {
137    /// The context in which the agent operates, by default this is the `DefaultContext`.
138    pub fn context(&mut self, context: impl AgentContext + 'static) -> &mut AgentBuilder
139    where
140        Self: Clone,
141    {
142        self.context = Some(Arc::new(context) as Arc<dyn AgentContext>);
143        self
144    }
145
146    /// Disable the system prompt.
147    pub fn no_system_prompt(&mut self) -> &mut Self {
148        self.system_prompt = Some(None);
149
150        self
151    }
152
153    /// Add a hook to the agent.
154    pub fn add_hook(&mut self, hook: Hook) -> &mut Self {
155        let hooks = self.hooks.get_or_insert_with(Vec::new);
156        hooks.push(hook);
157
158        self
159    }
160
161    /// Add a hook that runs once, before all completions. Even if the agent is paused and resumed,
162    /// before all will not trigger again.
163    pub fn before_all(&mut self, hook: impl BeforeAllFn + 'static) -> &mut Self {
164        self.add_hook(Hook::BeforeAll(Box::new(hook)))
165    }
166
167    /// Add a hook that runs once, when the agent starts. This hook also runs if the agent stopped
168    /// and then starts again. The hook runs after any `before_all` hooks and before the
169    /// `before_completion` hooks.
170    pub fn on_start(&mut self, hook: impl OnStartFn + 'static) -> &mut Self {
171        self.add_hook(Hook::OnStart(Box::new(hook)))
172    }
173
174    /// Add a hook that runs before each completion.
175    pub fn before_completion(&mut self, hook: impl BeforeCompletionFn + 'static) -> &mut Self {
176        self.add_hook(Hook::BeforeCompletion(Box::new(hook)))
177    }
178
179    /// Add a hook that runs after each tool. The `Result<ToolOutput, ToolError>` is provided
180    /// as mut, so the tool output can be fully modified.
181    ///
182    /// The `ToolOutput` also references the original `ToolCall`, allowing you to match at runtime
183    /// what tool to interact with.
184    pub fn after_tool(&mut self, hook: impl AfterToolFn + 'static) -> &mut Self {
185        self.add_hook(Hook::AfterTool(Box::new(hook)))
186    }
187
188    /// Add a hook that runs before each tool. Yields an immutable reference to the `ToolCall`.
189    pub fn before_tool(&mut self, hook: impl BeforeToolFn + 'static) -> &mut Self {
190        self.add_hook(Hook::BeforeTool(Box::new(hook)))
191    }
192
193    /// Add a hook that runs after each completion, before tool invocation and/or new messages.
194    pub fn after_completion(&mut self, hook: impl AfterCompletionFn + 'static) -> &mut Self {
195        self.add_hook(Hook::AfterCompletion(Box::new(hook)))
196    }
197
198    /// Add a hook that runs after each completion, after tool invocations, right before a new loop
199    /// might start
200    pub fn after_each(&mut self, hook: impl AfterEachFn + 'static) -> &mut Self {
201        self.add_hook(Hook::AfterEach(Box::new(hook)))
202    }
203
204    /// Add a hook that runs when a new message is added to the context. Note that each tool adds a
205    /// separate message.
206    pub fn on_new_message(&mut self, hook: impl MessageHookFn + 'static) -> &mut Self {
207        self.add_hook(Hook::OnNewMessage(Box::new(hook)))
208    }
209
210    /// Set the LLM for the agent. An LLM must implement the `ChatCompletion` trait.
211    pub fn llm<LLM: ChatCompletion + Clone + 'static>(&mut self, llm: &LLM) -> &mut Self {
212        let boxed: Box<dyn ChatCompletion> = Box::new(llm.clone()) as Box<dyn ChatCompletion>;
213
214        self.llm = Some(boxed);
215        self
216    }
217
218    /// Define the available tools for the agent. Tools must implement the `Tool` trait.
219    ///
220    /// See the [tool attribute macro](`swiftide_macros::tool`) and the [tool derive
221    /// macro](`swiftide_macros::Tool`) for easy ways to create (many) tools.
222    pub fn tools<TOOL, I: IntoIterator<Item = TOOL>>(&mut self, tools: I) -> &mut Self
223    where
224        TOOL: Into<Box<dyn Tool>>,
225    {
226        self.tools = Some(
227            tools
228                .into_iter()
229                .map(Into::into)
230                .chain(Agent::default_tools())
231                .collect(),
232        );
233        self
234    }
235}
236
237impl Agent {
238    /// Build a new agent
239    pub fn builder() -> AgentBuilder {
240        AgentBuilder::default()
241    }
242}
243
244impl Agent {
245    /// Default tools for the agent that it always includes
246    fn default_tools() -> HashSet<Box<dyn Tool>> {
247        HashSet::from([Box::new(Stop::default()) as Box<dyn Tool>])
248    }
249
250    /// Run the agent with a user message. The agent will loop completions, make tool calls, until
251    /// no new messages are available.
252    #[tracing::instrument(skip_all, name = "agent.query")]
253    pub async fn query(&mut self, query: impl Into<String> + std::fmt::Debug) -> Result<()> {
254        self.run_agent(Some(query.into()), false).await
255    }
256
257    /// Run the agent with a user message once.
258    #[tracing::instrument(skip_all, name = "agent.query_once")]
259    pub async fn query_once(&mut self, query: impl Into<String> + std::fmt::Debug) -> Result<()> {
260        self.run_agent(Some(query.into()), true).await
261    }
262
263    /// Run the agent with without user message. The agent will loop completions, make tool calls,
264    /// until no new messages are available.
265    #[tracing::instrument(skip_all, name = "agent.run")]
266    pub async fn run(&mut self) -> Result<()> {
267        self.run_agent(None, false).await
268    }
269
270    /// Run the agent with without user message. The agent will loop completions, make tool calls,
271    /// until
272    #[tracing::instrument(skip_all, name = "agent.run_once")]
273    pub async fn run_once(&mut self) -> Result<()> {
274        self.run_agent(None, true).await
275    }
276
277    /// Retrieve the message history of the agent
278    pub async fn history(&self) -> Vec<ChatMessage> {
279        self.context.history().await
280    }
281
282    async fn run_agent(&mut self, maybe_query: Option<String>, just_once: bool) -> Result<()> {
283        if self.state.is_running() {
284            anyhow::bail!("Agent is already running");
285        }
286
287        if self.state.is_pending() {
288            if let Some(system_prompt) = &self.system_prompt {
289                self.context
290                    .add_messages(vec![ChatMessage::System(system_prompt.render().await?)])
291                    .await;
292            }
293            for hook in self.hooks_by_type(HookTypes::BeforeAll) {
294                if let Hook::BeforeAll(hook) = hook {
295                    let span = tracing::info_span!(
296                        "hook",
297                        "otel.name" = format!("hook.{}", HookTypes::AfterTool)
298                    );
299                    tracing::info!("Calling {} hook", HookTypes::AfterTool);
300                    hook(self).instrument(span.or_current()).await?;
301                }
302            }
303        }
304
305        for hook in self.hooks_by_type(HookTypes::OnStart) {
306            if let Hook::OnStart(hook) = hook {
307                let span = tracing::info_span!(
308                    "hook",
309                    "otel.name" = format!("hook.{}", HookTypes::OnStart)
310                );
311                tracing::info!("Calling {} hook", HookTypes::OnStart);
312                hook(self).instrument(span.or_current()).await?;
313            }
314        }
315
316        self.state = state::State::Running;
317
318        if let Some(query) = maybe_query {
319            self.context.add_message(ChatMessage::User(query)).await;
320        }
321
322        let mut loop_counter = 0;
323
324        while let Some(messages) = self.context.next_completion().await {
325            if let Some(limit) = self.limit {
326                if loop_counter >= limit {
327                    tracing::warn!("Agent loop limit reached");
328                    break;
329                }
330            }
331            let result = self.run_completions(&messages).await;
332
333            if let Err(err) = result {
334                self.stop();
335                tracing::error!(error = ?err, "Agent stopped with error {err}");
336                return Err(err);
337            }
338
339            if just_once || self.state.is_stopped() {
340                break;
341            }
342            loop_counter += 1;
343        }
344
345        // If there are no new messages, ensure we update our state
346        self.stop();
347
348        Ok(())
349    }
350
351    #[tracing::instrument(skip_all, err)]
352    async fn run_completions(&mut self, messages: &[ChatMessage]) -> Result<()> {
353        debug!(
354            "Running completion for agent with {} messages",
355            messages.len()
356        );
357
358        let mut chat_completion_request = ChatCompletionRequest::builder()
359            .messages(messages)
360            .tools_spec(
361                self.tools
362                    .iter()
363                    .map(swiftide_core::Tool::tool_spec)
364                    .collect::<HashSet<_>>(),
365            )
366            .build()?;
367
368        for hook in self.hooks_by_type(HookTypes::BeforeCompletion) {
369            if let Hook::BeforeCompletion(hook) = hook {
370                let span = tracing::info_span!(
371                    "hook",
372                    "otel.name" = format!("hook.{}", HookTypes::BeforeCompletion)
373                );
374                tracing::info!("Calling {} hook", HookTypes::BeforeCompletion);
375                hook(&*self, &mut chat_completion_request)
376                    .instrument(span.or_current())
377                    .await?;
378            }
379        }
380
381        debug!(
382            "Calling LLM with the following new messages:\n {}",
383            self.context
384                .current_new_messages()
385                .await
386                .iter()
387                .map(ToString::to_string)
388                .collect::<Vec<_>>()
389                .join(",\n")
390        );
391
392        let mut response = self.llm.complete(&chat_completion_request).await?;
393
394        for hook in self.hooks_by_type(HookTypes::AfterCompletion) {
395            if let Hook::AfterCompletion(hook) = hook {
396                let span = tracing::info_span!(
397                    "hook",
398                    "otel.name" = format!("hook.{}", HookTypes::AfterCompletion)
399                );
400                tracing::info!("Calling {} hook", HookTypes::AfterCompletion);
401                hook(&*self, &mut response)
402                    .instrument(span.or_current())
403                    .await?;
404            }
405        }
406        self.add_message(ChatMessage::Assistant(
407            response.message,
408            response.tool_calls.clone(),
409        ))
410        .await?;
411
412        if let Some(tool_calls) = response.tool_calls {
413            self.invoke_tools(tool_calls).await?;
414        };
415
416        for hook in self.hooks_by_type(HookTypes::AfterEach) {
417            if let Hook::AfterEach(hook) = hook {
418                let span = tracing::info_span!(
419                    "hook",
420                    "otel.name" = format!("hook.{}", HookTypes::AfterEach)
421                );
422                tracing::info!("Calling {} hook", HookTypes::AfterEach);
423                hook(&*self).instrument(span.or_current()).await?;
424            }
425        }
426
427        Ok(())
428    }
429
430    async fn invoke_tools(&mut self, tool_calls: Vec<ToolCall>) -> Result<()> {
431        debug!("LLM returned tool calls: {:?}", tool_calls);
432
433        let mut handles = vec![];
434        for tool_call in tool_calls {
435            let Some(tool) = self.find_tool_by_name(tool_call.name()) else {
436                tracing::warn!("Tool {} not found", tool_call.name());
437                continue;
438            };
439            tracing::info!("Calling tool `{}`", tool_call.name());
440
441            let tool_args = tool_call.args().map(String::from);
442            let context: Arc<dyn AgentContext> = Arc::clone(&self.context);
443
444            for hook in self.hooks_by_type(HookTypes::BeforeTool) {
445                if let Hook::BeforeTool(hook) = hook {
446                    let span = tracing::info_span!(
447                        "hook",
448                        "otel.name" = format!("hook.{}", HookTypes::BeforeTool)
449                    );
450                    tracing::info!("Calling {} hook", HookTypes::BeforeTool);
451                    hook(&*self, &tool_call)
452                        .instrument(span.or_current())
453                        .await?;
454                }
455            }
456
457            let tool_span = tracing::info_span!(
458                "tool",
459                "otel.name" = format!("tool.{}", tool.name().as_ref())
460            );
461
462            let handle = tokio::spawn(async move {
463                    let tool_args = ArgPreprocessor::preprocess(tool_args.as_deref());
464                    let output = tool.invoke(&*context, tool_args.as_deref()).await.map_err(|e| { tracing::error!(error = %e, "Failed tool call"); e })?;
465
466                    tracing::debug!(output = output.to_string(), args = ?tool_args, tool_name = tool.name().as_ref(), "Completed tool call");
467
468                    Ok(output)
469                }.instrument(tool_span.or_current()));
470
471            handles.push((handle, tool_call));
472        }
473
474        for (handle, tool_call) in handles {
475            let mut output = handle.await?;
476
477            // Invoking hooks feels too verbose and repetitive
478            for hook in self.hooks_by_type(HookTypes::AfterTool) {
479                if let Hook::AfterTool(hook) = hook {
480                    let span = tracing::info_span!(
481                        "hook",
482                        "otel.name" = format!("hook.{}", HookTypes::AfterTool)
483                    );
484                    tracing::info!("Calling {} hook", HookTypes::AfterTool);
485                    hook(&*self, &tool_call, &mut output)
486                        .instrument(span.or_current())
487                        .await?;
488                }
489            }
490
491            if let Err(error) = output {
492                let stop = self.tool_calls_over_limit(&tool_call);
493                if stop {
494                    tracing::error!(
495                        "Tool call failed, retry limit reached, stopping agent: {err}",
496                        err = error
497                    );
498                } else {
499                    tracing::warn!(
500                        error = error.to_string(),
501                        tool_call = ?tool_call,
502                        "Tool call failed, retrying",
503                    );
504                }
505                self.add_message(ChatMessage::ToolOutput(
506                    tool_call,
507                    ToolOutput::Fail(error.to_string()),
508                ))
509                .await?;
510                if stop {
511                    self.stop();
512                    return Err(error.into());
513                }
514                continue;
515            }
516
517            let output = output?;
518            self.handle_control_tools(&output);
519            self.add_message(ChatMessage::ToolOutput(tool_call, output))
520                .await?;
521        }
522
523        Ok(())
524    }
525
526    fn hooks_by_type(&self, hook_type: HookTypes) -> Vec<&Hook> {
527        self.hooks
528            .iter()
529            .filter(|h| hook_type == (*h).into())
530            .collect()
531    }
532
533    fn find_tool_by_name(&self, tool_name: &str) -> Option<Box<dyn Tool>> {
534        self.tools
535            .iter()
536            .find(|tool| tool.name() == tool_name)
537            .cloned()
538    }
539
540    // Handle any tool specific output (e.g. stop)
541    fn handle_control_tools(&mut self, output: &ToolOutput) {
542        if let ToolOutput::Stop = output {
543            tracing::warn!("Stop tool called, stopping agent");
544            self.stop();
545        }
546    }
547
548    fn tool_calls_over_limit(&mut self, tool_call: &ToolCall) -> bool {
549        let mut s = DefaultHasher::new();
550        tool_call.hash(&mut s);
551        let hash = s.finish();
552
553        if let Some(retries) = self.tool_retries_counter.get_mut(&hash) {
554            let val = *retries >= self.tool_retry_limit;
555            *retries += 1;
556            val
557        } else {
558            self.tool_retries_counter.insert(hash, 1);
559            false
560        }
561    }
562
563    #[tracing::instrument(skip_all, fields(message = message.to_string()))]
564    async fn add_message(&self, mut message: ChatMessage) -> Result<()> {
565        for hook in self.hooks_by_type(HookTypes::OnNewMessage) {
566            if let Hook::OnNewMessage(hook) = hook {
567                let span = tracing::info_span!(
568                    "hook",
569                    "otel.name" = format!("hook.{}", HookTypes::OnNewMessage)
570                );
571                if let Err(err) = hook(self, &mut message).instrument(span.or_current()).await {
572                    tracing::error!(
573                        "Error in {hooktype} hook: {err}",
574                        hooktype = HookTypes::OnNewMessage,
575                    );
576                }
577            }
578        }
579        self.context.add_message(message).await;
580        Ok(())
581    }
582
583    /// Tell the agent to stop. It will finish it's current loop and then stop.
584    pub fn stop(&mut self) {
585        self.state = state::State::Stopped;
586    }
587
588    /// Access the agent's context
589    pub fn context(&self) -> &dyn AgentContext {
590        &self.context
591    }
592
593    /// The agent is still running
594    pub fn is_running(&self) -> bool {
595        self.state.is_running()
596    }
597
598    /// The agent stopped
599    pub fn is_stopped(&self) -> bool {
600        self.state.is_stopped()
601    }
602
603    /// The agent has not (ever) started
604    pub fn is_pending(&self) -> bool {
605        self.state.is_pending()
606    }
607}
608
609#[cfg(test)]
610mod tests {
611
612    use serde::ser::Error;
613    use swiftide_core::chat_completion::errors::ToolError;
614    use swiftide_core::chat_completion::{ChatCompletionResponse, ToolCall};
615    use swiftide_core::test_utils::MockChatCompletion;
616
617    use super::*;
618    use crate::{
619        assistant, chat_request, chat_response, summary, system, tool_failed, tool_output, user,
620    };
621
622    use crate::test_utils::{MockHook, MockTool};
623
624    #[test_log::test(tokio::test)]
625    async fn test_agent_builder_defaults() {
626        // Create a prompt
627        let mock_llm = MockChatCompletion::new();
628
629        // Build the agent
630        let agent = Agent::builder().llm(&mock_llm).build().unwrap();
631
632        // Check that the context is the default context
633
634        // Check that the default tools are added
635        assert!(agent.find_tool_by_name("stop").is_some());
636
637        // Check it does not allow duplicates
638        let agent = Agent::builder()
639            .tools([Stop::default(), Stop::default()])
640            .llm(&mock_llm)
641            .build()
642            .unwrap();
643
644        assert_eq!(agent.tools.len(), 1);
645
646        // It should include the default tool if a different tool is provided
647        let agent = Agent::builder()
648            .tools([MockTool::new("mock_tool")])
649            .llm(&mock_llm)
650            .build()
651            .unwrap();
652
653        assert_eq!(agent.tools.len(), 2);
654        assert!(agent.find_tool_by_name("mock_tool").is_some());
655        assert!(agent.find_tool_by_name("stop").is_some());
656
657        assert!(agent.context().history().await.is_empty());
658    }
659
660    #[test_log::test(tokio::test)]
661    async fn test_agent_tool_calling_loop() {
662        let prompt = "Write a poem";
663        let mock_llm = MockChatCompletion::new();
664        let mock_tool = MockTool::new("mock_tool");
665
666        let chat_request = chat_request! {
667            user!("Write a poem");
668
669            tools = [mock_tool.clone()]
670        };
671
672        let mock_tool_response = chat_response! {
673            "Roses are red";
674            tool_calls = ["mock_tool"]
675
676        };
677
678        mock_llm.expect_complete(chat_request.clone(), Ok(mock_tool_response));
679
680        let chat_request = chat_request! {
681            user!("Write a poem"),
682            assistant!("Roses are red", ["mock_tool"]),
683            tool_output!("mock_tool", "Great!");
684
685            tools = [mock_tool.clone()]
686        };
687
688        let stop_response = chat_response! {
689            "Roses are red";
690            tool_calls = ["stop"]
691        };
692
693        mock_llm.expect_complete(chat_request, Ok(stop_response));
694        mock_tool.expect_invoke_ok("Great!".into(), None);
695
696        let mut agent = Agent::builder()
697            .tools([mock_tool])
698            .llm(&mock_llm)
699            .no_system_prompt()
700            .build()
701            .unwrap();
702
703        agent.query(prompt).await.unwrap();
704    }
705
706    #[test_log::test(tokio::test)]
707    async fn test_agent_tool_run_once() {
708        let prompt = "Write a poem";
709        let mock_llm = MockChatCompletion::new();
710        let mock_tool = MockTool::default();
711
712        let chat_request = chat_request! {
713            system!("My system prompt"),
714            user!("Write a poem");
715
716            tools = [mock_tool.clone()]
717        };
718
719        let mock_tool_response = chat_response! {
720            "Roses are red";
721            tool_calls = ["mock_tool"]
722
723        };
724
725        mock_tool.expect_invoke_ok("Great!".into(), None);
726        mock_llm.expect_complete(chat_request.clone(), Ok(mock_tool_response));
727
728        let mut agent = Agent::builder()
729            .tools([mock_tool])
730            .system_prompt("My system prompt")
731            .llm(&mock_llm)
732            .build()
733            .unwrap();
734
735        agent.query_once(prompt).await.unwrap();
736    }
737
738    #[test_log::test(tokio::test(flavor = "multi_thread"))]
739    async fn test_multiple_tool_calls() {
740        let prompt = "Write a poem";
741        let mock_llm = MockChatCompletion::new();
742        let mock_tool = MockTool::new("mock_tool1");
743        let mock_tool2 = MockTool::new("mock_tool2");
744
745        let chat_request = chat_request! {
746            system!("My system prompt"),
747            user!("Write a poem");
748
749
750
751            tools = [mock_tool.clone(), mock_tool2.clone()]
752        };
753
754        let mock_tool_response = chat_response! {
755            "Roses are red";
756
757            tool_calls = ["mock_tool1", "mock_tool2"]
758
759        };
760
761        dbg!(&chat_request);
762        mock_tool.expect_invoke_ok("Great!".into(), None);
763        mock_tool2.expect_invoke_ok("Great!".into(), None);
764        mock_llm.expect_complete(chat_request.clone(), Ok(mock_tool_response));
765
766        let chat_request = chat_request! {
767            system!("My system prompt"),
768            user!("Write a poem"),
769            assistant!("Roses are red", ["mock_tool1", "mock_tool2"]),
770            tool_output!("mock_tool1", "Great!"),
771            tool_output!("mock_tool2", "Great!");
772
773            tools = [mock_tool.clone(), mock_tool2.clone()]
774        };
775
776        let mock_tool_response = chat_response! {
777            "Ok!";
778
779            tool_calls = ["stop"]
780
781        };
782
783        mock_llm.expect_complete(chat_request, Ok(mock_tool_response));
784
785        let mut agent = Agent::builder()
786            .tools([mock_tool, mock_tool2])
787            .system_prompt("My system prompt")
788            .llm(&mock_llm)
789            .build()
790            .unwrap();
791
792        agent.query(prompt).await.unwrap();
793    }
794
795    #[test_log::test(tokio::test)]
796    async fn test_agent_state_machine() {
797        let prompt = "Write a poem";
798        let mock_llm = MockChatCompletion::new();
799
800        let chat_request = chat_request! {
801            user!("Write a poem");
802            tools = []
803        };
804        let mock_tool_response = chat_response! {
805            "Roses are red";
806            tool_calls = []
807        };
808
809        mock_llm.expect_complete(chat_request.clone(), Ok(mock_tool_response));
810        let mut agent = Agent::builder()
811            .llm(&mock_llm)
812            .no_system_prompt()
813            .build()
814            .unwrap();
815
816        // Agent has never run and is pending
817        assert!(agent.state.is_pending());
818        agent.query_once(prompt).await.unwrap();
819
820        // Agent is stopped, there might be more messages
821        assert!(agent.state.is_stopped());
822    }
823
824    #[test_log::test(tokio::test)]
825    async fn test_summary() {
826        let prompt = "Write a poem";
827        let mock_llm = MockChatCompletion::new();
828
829        let mock_tool_response = chat_response! {
830            "Roses are red";
831            tool_calls = []
832
833        };
834
835        let expected_chat_request = chat_request! {
836            system!("My system prompt"),
837            user!("Write a poem");
838
839            tools = []
840        };
841
842        mock_llm.expect_complete(expected_chat_request, Ok(mock_tool_response.clone()));
843
844        let mut agent = Agent::builder()
845            .system_prompt("My system prompt")
846            .llm(&mock_llm)
847            .build()
848            .unwrap();
849
850        agent.query_once(prompt).await.unwrap();
851
852        agent
853            .context
854            .add_message(ChatMessage::new_summary("Summary"))
855            .await;
856
857        let expected_chat_request = chat_request! {
858            system!("My system prompt"),
859            summary!("Summary"),
860            user!("Write another poem");
861            tools = []
862        };
863        mock_llm.expect_complete(expected_chat_request, Ok(mock_tool_response.clone()));
864
865        agent.query_once("Write another poem").await.unwrap();
866
867        agent
868            .context
869            .add_message(ChatMessage::new_summary("Summary 2"))
870            .await;
871
872        let expected_chat_request = chat_request! {
873            system!("My system prompt"),
874            summary!("Summary 2"),
875            user!("Write a third poem");
876            tools = []
877        };
878        mock_llm.expect_complete(expected_chat_request, Ok(mock_tool_response));
879
880        agent.query_once("Write a third poem").await.unwrap();
881    }
882
883    #[test_log::test(tokio::test)]
884    async fn test_agent_hooks() {
885        let mock_before_all = MockHook::new("before_all").expect_calls(1).to_owned();
886        let mock_on_start_fn = MockHook::new("on_start").expect_calls(1).to_owned();
887        let mock_before_completion = MockHook::new("before_completion")
888            .expect_calls(2)
889            .to_owned();
890        let mock_after_completion = MockHook::new("after_completion").expect_calls(2).to_owned();
891        let mock_after_each = MockHook::new("after_each").expect_calls(2).to_owned();
892        let mock_on_message = MockHook::new("on_message").expect_calls(4).to_owned();
893
894        // Once for mock tool and once for stop
895        let mock_before_tool = MockHook::new("before_tool").expect_calls(2).to_owned();
896        let mock_after_tool = MockHook::new("after_tool").expect_calls(2).to_owned();
897
898        let prompt = "Write a poem";
899        let mock_llm = MockChatCompletion::new();
900        let mock_tool = MockTool::default();
901
902        let chat_request = chat_request! {
903            user!("Write a poem");
904
905            tools = [mock_tool.clone()]
906        };
907
908        let mock_tool_response = chat_response! {
909            "Roses are red";
910            tool_calls = ["mock_tool"]
911
912        };
913
914        mock_llm.expect_complete(chat_request.clone(), Ok(mock_tool_response));
915
916        let chat_request = chat_request! {
917            user!("Write a poem"),
918            assistant!("Roses are red", ["mock_tool"]),
919            tool_output!("mock_tool", "Great!");
920
921            tools = [mock_tool.clone()]
922        };
923
924        let stop_response = chat_response! {
925            "Roses are red";
926            tool_calls = ["stop"]
927        };
928
929        mock_llm.expect_complete(chat_request, Ok(stop_response));
930        mock_tool.expect_invoke_ok("Great!".into(), None);
931
932        let mut agent = Agent::builder()
933            .tools([mock_tool])
934            .llm(&mock_llm)
935            .no_system_prompt()
936            .before_all(mock_before_all.hook_fn())
937            .on_start(mock_on_start_fn.on_start_fn())
938            .before_completion(mock_before_completion.before_completion_fn())
939            .before_tool(mock_before_tool.before_tool_fn())
940            .after_completion(mock_after_completion.after_completion_fn())
941            .after_tool(mock_after_tool.after_tool_fn())
942            .after_each(mock_after_each.hook_fn())
943            .on_new_message(mock_on_message.message_hook_fn())
944            .build()
945            .unwrap();
946
947        agent.query(prompt).await.unwrap();
948    }
949
950    #[test_log::test(tokio::test)]
951    async fn test_agent_loop_limit() {
952        let prompt = "Generate content"; // Example prompt
953        let mock_llm = MockChatCompletion::new();
954        let mock_tool = MockTool::new("mock_tool");
955
956        let chat_request = chat_request! {
957            user!(prompt);
958            tools = [mock_tool.clone()]
959        };
960        mock_tool.expect_invoke_ok("Great!".into(), None);
961
962        let mock_tool_response = chat_response! {
963            "Some response";
964            tool_calls = ["mock_tool"]
965        };
966
967        // Set expectations for the mock LLM responses
968        mock_llm.expect_complete(chat_request.clone(), Ok(mock_tool_response.clone()));
969
970        // // Response for terminating the loop
971        let stop_response = chat_response! {
972            "Final response";
973            tool_calls = ["stop"]
974        };
975
976        mock_llm.expect_complete(chat_request, Ok(stop_response));
977
978        let mut agent = Agent::builder()
979            .tools([mock_tool])
980            .llm(&mock_llm)
981            .no_system_prompt()
982            .limit(1) // Setting the loop limit to 1
983            .build()
984            .unwrap();
985
986        // Run the agent
987        agent.query(prompt).await.unwrap();
988
989        // Assert that the remaining message is still in the queue
990        let remaining = mock_llm.expectations.lock().unwrap().pop();
991        assert!(remaining.is_some());
992
993        // Assert that the agent is stopped after reaching the loop limit
994        assert!(agent.is_stopped());
995    }
996
997    #[test_log::test(tokio::test)]
998    async fn test_tool_retry_mechanism() {
999        let prompt = "Execute my tool";
1000        let mock_llm = MockChatCompletion::new();
1001        let mock_tool = MockTool::new("retry_tool");
1002
1003        // Configure mock tool to fail twice. First time is fed back to the LLM, second time is an
1004        // error
1005        mock_tool.expect_invoke(
1006            Err(ToolError::WrongArguments(serde_json::Error::custom(
1007                "missing `query`",
1008            ))),
1009            None,
1010        );
1011        mock_tool.expect_invoke(
1012            Err(ToolError::WrongArguments(serde_json::Error::custom(
1013                "missing `query`",
1014            ))),
1015            None,
1016        );
1017
1018        let chat_request = chat_request! {
1019            user!(prompt);
1020            tools = [mock_tool.clone()]
1021        };
1022        let retry_response = chat_response! {
1023            "First failing attempt";
1024            tool_calls = ["retry_tool"]
1025        };
1026        mock_llm.expect_complete(chat_request.clone(), Ok(retry_response));
1027
1028        let chat_request = chat_request! {
1029            user!(prompt),
1030            assistant!("First failing attempt", ["retry_tool"]),
1031            tool_failed!("retry_tool", "arguments for tool failed to parse: missing `query`");
1032
1033            tools = [mock_tool.clone()]
1034        };
1035        let will_fail_response = chat_response! {
1036            "Finished execution";
1037            tool_calls = ["retry_tool"]
1038        };
1039        mock_llm.expect_complete(chat_request.clone(), Ok(will_fail_response));
1040
1041        let mut agent = Agent::builder()
1042            .tools([mock_tool])
1043            .llm(&mock_llm)
1044            .no_system_prompt()
1045            .tool_retry_limit(1) // The test relies on a limit of 2 retries.
1046            .build()
1047            .unwrap();
1048
1049        // Run the agent
1050        let result = agent.query(prompt).await;
1051
1052        assert!(result.is_err());
1053        assert!(result.unwrap_err().to_string().contains("missing `query`"));
1054        assert!(agent.is_stopped());
1055    }
1056}
swiftide_agents/agent.rs

swiftide_agents/
agent.rs
