Files
optimclaw/tests/support_unit_tests.rs
T
b4b19738a8 Trajectory benchmarks and e2e trace test rig (#553)
* refactor: extract shared assertion helpers to support/assertions.rs

Move 5 assertion helpers from e2e_spot_checks.rs to a shared module.
Add assert_all_tools_succeeded and assert_tool_succeeded for eliminating
false positives in E2E tests.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* feat: add tool output capture via tool_results() accessor

Extract (name, preview) from ToolResult status events in TestChannel
and TestRig, enabling content assertions on tool outputs.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* fix: correct tool parameters in 3 broken trace fixtures

- tool_time.json: add missing "operation": "now" for time tool
- robust_correct_tool.json: same fix
- memory_full_cycle.json: change "path" to "target" for memory_write

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* fix: add tool success and output assertions to eliminate false positives

Every E2E test that exercises tools now calls assert_all_tools_succeeded.
Added tool output content assertions where tool results are predictable
(time year, read_file content, memory_read content).

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* feat: capture per-tool timing from ToolStarted/ToolCompleted events

Record Instant on ToolStarted and compute elapsed duration on
ToolCompleted, wiring real timing data into collect_metrics() instead
of hardcoded zeros.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* refactor: add RAII CleanupGuard for temp file/dir cleanup in tests

Replace manual cleanup_test_dir() calls and inline remove_file() with
Drop-based CleanupGuard that ensures cleanup even if a test panics.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* fix: add Drop impl and graceful shutdown for TestRig

Wrap agent_handle in Option so Drop can abort leaked tasks. Signal
the channel shutdown before aborting for future cooperative shutdown.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* fix: replace agent startup sleep with oneshot ready signal

Use a oneshot channel fired in Channel::start() instead of a fixed
100ms sleep, eliminating the race condition on slow systems.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* fix: replace fragile string-matching iteration limit with count-based detection

Use tool completion count vs max_tool_iterations instead of scanning
status messages for "iteration"/"limit" substrings.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* fix: use assert_all_tools_succeeded for memory_full_cycle test

Remove incorrect comment about memory_tree failing with empty path
(it actually succeeds). Omit empty path from fixture and use the
standard assert_all_tools_succeeded instead of per-tool assertions.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* refactor: promote benchmark metrics types to library code

Move TraceMetrics, ScenarioResult, RunResult, MetricDelta, and
compare_runs() from tests/support/metrics.rs to src/benchmark/metrics.rs.
Existing tests use re-export for backward compatibility.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* feat: add Scenario and Criterion types for agent benchmarking

Scenario defines a task with input, success criteria, and resource
limits. Criterion is an enum of programmatic checks (tool_used,
response_contains, etc.) evaluated without LLM judgment.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* feat: add initial benchmark scenario suite (12 scenarios across 5 categories)

Scenarios cover tool_selection, tool_chaining, error_recovery,
efficiency, and memory_operations. All loaded from JSON with
deserialization validation test.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* feat: add benchmark runner with BenchChannel and InstrumentedLlm

BenchChannel is a minimal Channel implementation for benchmarks.
InstrumentedLlm wraps any LlmProvider to capture per-call metrics.
Runner creates a fresh agent per scenario, evaluates success criteria,
and produces RunResult with timing, token, and cost metrics.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* feat: add baseline management, reports, and benchmark entry point

- baseline.rs: load/save/promote benchmark results
- report.rs: format comparison reports with regression detection
- benchmark_runner.rs: integration test with real LLM (feature-gated)
- Add benchmark feature flag to Cargo.toml

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* style: apply cargo fmt to benchmark module

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* feat(benchmark): add multi-turn scenario types with setup, judge, ResponseNotContains

Add BenchScenario, Turn, TurnAssertions, JudgeConfig, ScenarioSetup,
WorkspaceSetup, SeedDocument types for multi-turn benchmark scenarios.
Add ResponseNotContains criterion variant. Add TurnAssertions::to_criteria()
converter for backward compat with existing evaluation engine.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* feat(benchmark): add JSON scenario loader with recursive discovery and tag filter

Add load_bench_scenarios() for the new BenchScenario format with recursive
directory traversal and tag-based filtering. Create 4 initial trajectory
scenarios across tool-selection, multi-turn, and efficiency categories.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* feat(benchmark): multi-turn runner with workspace seeding and per-turn metrics

Add run_bench_scenario() that loops over BenchScenario turns, seeds workspace
documents, collects per-turn metrics (tokens, tool calls, wall time), and
evaluates per-turn assertions. Add TurnMetrics to metrics.rs and
clear_for_next_turn() to BenchChannel.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* feat(benchmark): add LLM-as-judge scoring with prompt formatting and score parsing

Create judge.rs with format_judge_prompt, parse_judge_score, and judge_turn.
Wire into run_bench_scenario for turns with judge config -- scores below
min_score fail the turn.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* feat(benchmark): add CLI subcommand (ironclaw benchmark)

Add BenchmarkCommand with --tags, --scenario, --no-judge, --timeout,
--update-baseline flags. Wire into Command enum and main.rs dispatch.
Feature-gated behind benchmark flag.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* feat(benchmark): per-scenario JSON output with full trajectory

Add save_scenario_results() that writes per-scenario JSON files alongside
the run summary. Each scenario gets its own file with turn_metrics trajectory.
Update CLI to use new output format.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* feat(benchmark): add ToolRegistry::retain_only and wire tool filtering in scenarios

Add a retain_only() method to ToolRegistry that filters tools down to a
given allowlist. Wire this into run_bench_scenario() so that when a
scenario specifies a tools list in its setup, only those tools are
available during the benchmark run. Includes two tests for the new
method: one verifying filtering works and one verifying empty input
is a no-op.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* feat(benchmark): wire identity overrides into workspace before agent start

Add seed_identity() helper that writes identity files (IDENTITY.md,
USER.md, etc.) into the workspace before the agent starts, so that
workspace.system_prompt() picks them up. Wire it into
run_bench_scenario() after workspace seeding. Include a test that
verifies identity files are written and readable.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* feat(benchmark): add --parallel and --max-cost CLI flags

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* fix(benchmark): use feature-conditional snapshot names for CLI help tests

Prevents snapshot conflicts between default (no benchmark) and
all-features (with benchmark) builds by using separate snapshot names
per feature set.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* feat(benchmark): parallel execution with JoinSet and budget cap enforcement

Replace sequential loop in run_all_bench() with parallel execution using
JoinSet + semaphore when config.parallel > 1. Add budget cap enforcement
that skips remaining scenarios when max_total_cost_usd is exceeded.
Track skipped count in RunResult.skipped_scenarios and display it in
format_report().

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* feat(benchmark): add tool restriction and identity override test scenarios

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* chore: fix formatting for Phase 3

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* feat(benchmark): add SkillRegistry::retain_only and wire skill filtering in scenarios

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* feat(benchmark): add --json flag for machine-readable output

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* ci: add GitHub Actions benchmark workflow (manual trigger)

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* refactor(benchmark): remove in-tree benchmark harness, keep retain_only utilities

Move benchmark-specific code out of ironclaw in preparation for the
nearai/benchmarks trajectory adapter. This removes:

- src/benchmark/ (runner, scenarios, metrics, judge, report, etc.)
- src/cli/benchmark.rs and the Benchmark CLI subcommand
- benchmarks/ data directory (scenarios + trajectories)
- .github/workflows/benchmark.yml
- The "benchmark" Cargo feature flag

What remains:
- ToolRegistry::retain_only() and SkillRegistry::retain_only()
- Test support types (TraceMetrics, InstrumentedLlm) inlined into
  tests/support/ instead of re-exporting from the deleted module

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* docs: add README for LLM trace fixture format

Documents the trajectory JSON format, response types, request hints,
directory structure, and how to write new traces.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* feat(test): unify trace format around turns, add multi-turn support

Introduce TraceTurn type that groups user_input with LLM response steps,
making traces self-contained conversation trajectories. Add run_trace()
to TestRig for automatic multi-turn replay. Backward-compatible: flat
"steps" JSON is deserialized as a single turn transparently.

Includes all trace fixtures (spot, coverage, advanced), plan docs, and
new e2e tests for steering, error recovery, long chains, memory, and
prompt injection resilience.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* fix(test): fix CI failures after merging main

- Fix tool_json fixture: use "data" parameter (not "input") to match
  JsonTool schema
- Fix status_events test: remove assertion for "time" tool that isn't
  in the fixture (only "echo" calls are used)
- Allow dead_code in test support metrics/instrumented_llm modules
  (utilities for future benchmark tests)

[skip-regression-check]

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* Working on recording traces and testing them

* feat(test): add declarative expects to trace fixtures, split infra tests

Add TraceExpects struct with 9 optional assertion fields (response_contains,
tools_used, all_tools_succeeded, etc.) that can be declared in fixture JSON
instead of hand-written Rust. Add verify_expects() and run_recorded_trace()
so recorded trace tests become one-liners.

Split trace infra tests (deserialization, backward compat) into
tests/trace_format.rs which doesn't require the libsql feature gate.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* refactor(test): add expects to all trace fixtures, simplify e2e tests

Add declarative expects blocks to all 19 trace fixture JSONs across
spot/, coverage/, advanced/, and root directories. Update all 8 e2e
test files to use verify_trace_expects() / run_and_verify_trace(),
replacing ~270 lines of hand-written assertions with fixture-driven
verification.

Tests that check things beyond expects (file content on disk, metrics,
event ordering) keep those extra assertions alongside the declarative
ones.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* fix(test): adapt tests to AppBuilder refactor, fix formatting

Update test files to work with refactored TestRigBuilder that uses
AppBuilder::build_all() (removing with_tools/with_workspace methods).
Update telegram_check fixture to use tool_list instead of echo.
Fix cargo fmt issues in src/llm/mod.rs and src/llm/recording.rs.

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* refactor(test): deduplicate support unit tests into single binary

Support modules (assertions, cleanup, test_channel, test_rig, trace_llm)
had #[cfg(test)] mod tests blocks that were compiled and run 12 times —
once per e2e test binary that declares `mod support;`. Extracted all 29
support unit tests into a dedicated `tests/support_unit_tests.rs` so they
run exactly once.

[skip-regression-check]

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* style: fix trailing newlines in support files

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* refactor(test): unify trace types and fix recorded multi-turn replay

Import shared types (TraceStep, TraceResponse, TraceToolCall, RequestHint,
ExpectedToolResult, MemorySnapshotEntry, HttpExchange*) from
ironclaw::llm::recording instead of redefining them in trace_llm.rs.

Fix the flat-steps deserializer to split at UserInput boundaries into
multiple turns, instead of filtering them out and wrapping everything
into a single turn. This enables recorded multi-turn traces to be
replayed as proper multi-turn conversations via run_trace().

[skip-regression-check]

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* fix(test): fix CI failures - unused imports and missing struct fields

- Add #[allow(unused_imports)] on pub use re-exports in trace_llm.rs
  (types are re-exported for downstream test files, not used locally)
- Add `..` to ToolCompleted pattern in test_channel.rs to match new
  `error` and `parameters` fields

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* fix(test): fix CI failures after merging main

- Add missing `error` and `parameters` fields to ToolCompleted
  constructors in support_unit_tests.rs
- Add `..` to ToolCompleted pattern match in support_unit_tests.rs
- Add #[allow(dead_code)] to CleanupGuard, LlmTrace impl, and
  TraceLlm impl (only used behind #[cfg(feature = "libsql")])

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* Adding coverage running script

* fix(test): address review feedback on E2E test infrastructure

- Increase wait_for_responses polling to exponential backoff (50ms-500ms)
  and raise default timeout from 15s to 30s to reduce CI flakiness (#1)
- Strengthen prompt_injection_resilience test with positive safety layer
  assertion via has_safety_warnings(), enable injection_check (#2)
- Add assert_tool_order() helper and tools_order field in TraceExpects
  for verifying tool execution ordering in multi-step traces (#3)
- Document TraceLlm sequential-call assumption for concurrency (#6)
- Clean up CleanupGuard with PathKind enum instead of shotgun
  remove_file + remove_dir_all on every path (#8)
- Fix coverage.sh: default to --lib only, fix multi-filter syntax,
  add COV_ALL_TARGETS option
- Add coverage/ to .gitignore
- Remove planning docs from PR

[skip-regression-check]

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* fix: address PR review - use HashSet in retain_only, improve skill test

- Use HashSet for O(N+M) lookup in SkillRegistry::retain_only and
  ToolRegistry::retain_only instead of linear scan
- Strengthen test_retain_only_empty_is_noop in SkillRegistry to
  pre-populate with a skill before asserting the no-op behavior

[skip-regression-check]

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* fix(test): revert incorrect safety layer assertion in injection test

The safety layer sanitizes tool output, not user input. The injection
test sends a malicious user message with no tools called, so the safety
layer never fires. Reverted to the original test which correctly
validates the LLM refuses via trace expects. Also fixed case-sensitive
request hint ("ignore" -> "Ignore") to suppress noisy warning.

[skip-regression-check]

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* fix: clean stale profdata before coverage run

Adds `cargo llvm-cov clean` before each run to prevent
"mismatched data" warnings from stale instrumentation profiles.

[skip-regression-check]

Co-Authored-By: Claude Opus 4.6 <[email protected]>

* style: fix formatting in retain_only test

[skip-regression-check]

Co-Authored-By: Claude Opus 4.6 <[email protected]>

---------

Co-authored-by: Claude Opus 4.6 <[email protected]>
Co-authored-by: Illia Polosukhin <[email protected]>
2026-03-05 09:13:09 +00:00

726 lines
23 KiB
Rust

//! Unit tests for E2E test support modules.
//!
//! These tests live here (instead of inside `support/*.rs`) so they compile
//! and run exactly once, rather than being duplicated across every `e2e_*.rs`
//! test binary that declares `mod support;`.
mod support;
// ---------------------------------------------------------------------------
// assertions
// ---------------------------------------------------------------------------
mod assertions_tests {
use crate::support::assertions::*;
#[test]
fn all_tools_succeeded_passes_when_all_true() {
let completed = vec![("echo".to_string(), true), ("time".to_string(), true)];
assert_all_tools_succeeded(&completed);
}
#[test]
fn all_tools_succeeded_passes_on_empty() {
assert_all_tools_succeeded(&[]);
}
#[test]
#[should_panic(expected = "Expected all tools to succeed")]
fn all_tools_succeeded_panics_on_failure() {
let completed = vec![("echo".to_string(), true), ("shell".to_string(), false)];
assert_all_tools_succeeded(&completed);
}
#[test]
fn tool_succeeded_passes_when_present_and_true() {
let completed = vec![("echo".to_string(), true), ("time".to_string(), false)];
assert_tool_succeeded(&completed, "echo");
}
#[test]
#[should_panic(expected = "Expected 'echo' to complete successfully")]
fn tool_succeeded_panics_when_tool_missing() {
let completed = vec![("time".to_string(), true)];
assert_tool_succeeded(&completed, "echo");
}
#[test]
#[should_panic(expected = "Expected 'shell' to complete successfully")]
fn tool_succeeded_panics_when_tool_failed() {
let completed = vec![("shell".to_string(), false)];
assert_tool_succeeded(&completed, "shell");
}
#[test]
fn tool_order_passes_for_correct_order() {
let started: Vec<String> = vec!["write_file", "echo", "read_file"]
.into_iter()
.map(String::from)
.collect();
assert_tool_order(&started, &["write_file", "read_file"]);
}
#[test]
fn tool_order_passes_for_consecutive() {
let started: Vec<String> = vec!["write_file", "read_file"]
.into_iter()
.map(String::from)
.collect();
assert_tool_order(&started, &["write_file", "read_file"]);
}
#[test]
#[should_panic(expected = "assert_tool_order")]
fn tool_order_panics_for_wrong_order() {
let started: Vec<String> = vec!["read_file", "write_file"]
.into_iter()
.map(String::from)
.collect();
assert_tool_order(&started, &["write_file", "read_file"]);
}
#[test]
#[should_panic(expected = "assert_tool_order")]
fn tool_order_panics_for_missing_tool() {
let started: Vec<String> = vec!["echo".to_string()];
assert_tool_order(&started, &["echo", "write_file"]);
}
}
// ---------------------------------------------------------------------------
// cleanup
// ---------------------------------------------------------------------------
mod cleanup_tests {
use crate::support::cleanup::CleanupGuard;
#[test]
fn cleanup_guard_removes_file() {
let path = "/tmp/ironclaw_cleanup_guard_test.txt";
std::fs::write(path, "test").unwrap();
{
let _guard = CleanupGuard::new().file(path);
assert!(std::path::Path::new(path).exists());
}
assert!(!std::path::Path::new(path).exists());
}
#[test]
fn cleanup_guard_removes_dir() {
let dir = "/tmp/ironclaw_cleanup_guard_test_dir";
std::fs::create_dir_all(dir).unwrap();
std::fs::write(format!("{dir}/file.txt"), "test").unwrap();
{
let _guard = CleanupGuard::new().dir(dir);
assert!(std::path::Path::new(dir).exists());
}
assert!(!std::path::Path::new(dir).exists());
}
#[test]
fn cleanup_guard_file_does_not_remove_dir() {
let dir = "/tmp/ironclaw_cleanup_guard_file_not_dir";
std::fs::create_dir_all(dir).unwrap();
{
// Registering a directory path as .file() should not remove it
// (remove_file fails on directories).
let _guard = CleanupGuard::new().file(dir);
}
assert!(
std::path::Path::new(dir).exists(),
"dir should still exist when registered as file"
);
// Clean up manually.
let _ = std::fs::remove_dir_all(dir);
}
}
// ---------------------------------------------------------------------------
// test_channel
// ---------------------------------------------------------------------------
mod test_channel_tests {
use std::sync::Arc;
use std::time::Duration;
use crate::support::test_channel::TestChannel;
use ironclaw::channels::{Channel, IncomingMessage, OutgoingResponse, StatusUpdate};
#[tokio::test]
async fn send_and_receive_message() {
let channel = TestChannel::new();
let mut stream = channel.start().await.unwrap();
channel.send_message("hello world").await;
use futures::StreamExt;
let msg = stream.next().await.expect("stream should yield a message");
assert_eq!(msg.content, "hello world");
assert_eq!(msg.channel, "test");
assert_eq!(msg.user_id, "test-user");
}
#[tokio::test]
async fn captures_responses() {
let channel = TestChannel::new();
let incoming = IncomingMessage::new("test", "test-user", "hi");
channel
.respond(&incoming, OutgoingResponse::text("reply 1"))
.await
.unwrap();
channel
.respond(&incoming, OutgoingResponse::text("reply 2"))
.await
.unwrap();
let captured = channel.captured_responses();
assert_eq!(captured.len(), 2);
assert_eq!(captured[0].content, "reply 1");
assert_eq!(captured[1].content, "reply 2");
}
#[tokio::test]
async fn captures_status_events() {
let channel = TestChannel::new();
let metadata = serde_json::Value::Null;
channel
.send_status(
StatusUpdate::ToolStarted {
name: "echo".to_string(),
},
&metadata,
)
.await
.unwrap();
channel
.send_status(
StatusUpdate::ToolCompleted {
name: "echo".to_string(),
success: true,
error: None,
parameters: None,
},
&metadata,
)
.await
.unwrap();
let events = channel.captured_status_events();
assert_eq!(events.len(), 2);
assert!(matches!(&events[0], StatusUpdate::ToolStarted { name } if name == "echo"));
assert!(
matches!(&events[1], StatusUpdate::ToolCompleted { name, success, .. } if name == "echo" && *success)
);
}
#[tokio::test]
async fn tool_calls_started() {
let channel = TestChannel::new();
let metadata = serde_json::Value::Null;
channel
.send_status(
StatusUpdate::ToolStarted {
name: "memory_search".to_string(),
},
&metadata,
)
.await
.unwrap();
channel
.send_status(StatusUpdate::Thinking("hmm".to_string()), &metadata)
.await
.unwrap();
channel
.send_status(
StatusUpdate::ToolStarted {
name: "echo".to_string(),
},
&metadata,
)
.await
.unwrap();
let started = channel.tool_calls_started();
assert_eq!(started, vec!["memory_search", "echo"]);
}
#[tokio::test]
async fn tool_results() {
let channel = TestChannel::new();
channel
.send_status(
StatusUpdate::ToolResult {
name: "echo".to_string(),
preview: "hello world".to_string(),
},
&serde_json::Value::Null,
)
.await
.unwrap();
channel
.send_status(
StatusUpdate::ToolResult {
name: "time".to_string(),
preview: "{\"iso\": \"2026-03-03\"}".to_string(),
},
&serde_json::Value::Null,
)
.await
.unwrap();
let results = channel.tool_results();
assert_eq!(results.len(), 2);
assert_eq!(results[0].0, "echo");
assert_eq!(results[0].1, "hello world");
assert_eq!(results[1].0, "time");
assert!(results[1].1.contains("2026"));
}
#[tokio::test]
async fn wait_for_responses() {
let channel = TestChannel::new();
let responses = Arc::clone(&channel.responses);
tokio::spawn(async move {
tokio::time::sleep(Duration::from_millis(100)).await;
responses
.lock()
.await
.push(OutgoingResponse::text("delayed reply"));
});
let collected = channel.wait_for_responses(1, Duration::from_secs(2)).await;
assert_eq!(collected.len(), 1);
assert_eq!(collected[0].content, "delayed reply");
}
#[tokio::test]
async fn tool_timings() {
let channel = TestChannel::new();
channel
.send_status(
StatusUpdate::ToolStarted {
name: "echo".to_string(),
},
&serde_json::Value::Null,
)
.await
.unwrap();
tokio::time::sleep(Duration::from_millis(50)).await;
channel
.send_status(
StatusUpdate::ToolCompleted {
name: "echo".to_string(),
success: true,
error: None,
parameters: None,
},
&serde_json::Value::Null,
)
.await
.unwrap();
let timings = channel.tool_timings();
assert_eq!(timings.len(), 1);
assert_eq!(timings[0].0, "echo");
assert!(
timings[0].1 >= 40,
"Expected >= 40ms, got {}ms",
timings[0].1
);
}
}
// ---------------------------------------------------------------------------
// trace_llm
// ---------------------------------------------------------------------------
mod trace_llm_tests {
use crate::support::trace_llm::*;
use ironclaw::llm::{
ChatMessage, CompletionRequest, FinishReason, LlmProvider, ToolCompletionRequest,
};
fn text_step(content: &str, input_tokens: u32, output_tokens: u32) -> TraceStep {
TraceStep {
request_hint: None,
response: TraceResponse::Text {
content: content.to_string(),
input_tokens,
output_tokens,
},
expected_tool_results: Vec::new(),
}
}
fn tool_calls_step(calls: Vec<TraceToolCall>, input: u32, output: u32) -> TraceStep {
TraceStep {
request_hint: None,
response: TraceResponse::ToolCalls {
tool_calls: calls,
input_tokens: input,
output_tokens: output,
},
expected_tool_results: Vec::new(),
}
}
fn simple_tool_call(name: &str) -> TraceToolCall {
TraceToolCall {
id: format!("call_{name}"),
name: name.to_string(),
arguments: serde_json::json!({"key": "value"}),
}
}
fn make_request(user_msg: &str) -> ToolCompletionRequest {
ToolCompletionRequest::new(vec![ChatMessage::user(user_msg)], vec![])
}
fn make_completion_request(user_msg: &str) -> CompletionRequest {
CompletionRequest::new(vec![ChatMessage::user(user_msg)])
}
#[tokio::test]
async fn replays_text_response() {
let trace =
LlmTrace::single_turn("test-model", "hi", vec![text_step("Hello world", 100, 20)]);
let llm = TraceLlm::from_trace(trace);
let resp = llm.complete_with_tools(make_request("hi")).await.unwrap();
assert_eq!(resp.content.as_deref(), Some("Hello world"));
assert!(resp.tool_calls.is_empty());
assert_eq!(resp.input_tokens, 100);
assert_eq!(resp.output_tokens, 20);
assert_eq!(resp.finish_reason, FinishReason::Stop);
assert_eq!(llm.calls(), 1);
}
#[tokio::test]
async fn replays_tool_calls() {
let trace = LlmTrace::single_turn(
"test-model",
"search memory",
vec![tool_calls_step(
vec![simple_tool_call("memory_search")],
80,
15,
)],
);
let llm = TraceLlm::from_trace(trace);
let resp = llm
.complete_with_tools(make_request("search memory"))
.await
.unwrap();
assert!(resp.content.is_none());
assert_eq!(resp.tool_calls.len(), 1);
assert_eq!(resp.tool_calls[0].name, "memory_search");
assert_eq!(resp.tool_calls[0].id, "call_memory_search");
assert_eq!(
resp.tool_calls[0].arguments,
serde_json::json!({"key": "value"})
);
assert_eq!(resp.input_tokens, 80);
assert_eq!(resp.output_tokens, 15);
assert_eq!(resp.finish_reason, FinishReason::ToolUse);
}
#[tokio::test]
async fn advances_through_steps() {
let trace = LlmTrace::single_turn(
"test-model",
"do something",
vec![
tool_calls_step(vec![simple_tool_call("echo")], 50, 10),
text_step("Done!", 60, 5),
],
);
let llm = TraceLlm::from_trace(trace);
let resp1 = llm
.complete_with_tools(make_request("do something"))
.await
.unwrap();
assert_eq!(resp1.tool_calls.len(), 1);
assert_eq!(resp1.tool_calls[0].name, "echo");
assert_eq!(llm.calls(), 1);
let resp2 = llm
.complete_with_tools(make_request("continue"))
.await
.unwrap();
assert_eq!(resp2.content.as_deref(), Some("Done!"));
assert!(resp2.tool_calls.is_empty());
assert_eq!(llm.calls(), 2);
}
#[tokio::test]
async fn errors_when_exhausted() {
let trace =
LlmTrace::single_turn("test-model", "first", vec![text_step("only once", 10, 5)]);
let llm = TraceLlm::from_trace(trace);
let resp1 = llm.complete_with_tools(make_request("first")).await;
assert!(resp1.is_ok());
let resp2 = llm.complete_with_tools(make_request("second")).await;
assert!(resp2.is_err());
let err = resp2.unwrap_err();
let err_msg = err.to_string();
assert!(
err_msg.contains("exhausted"),
"Expected 'exhausted' in error: {err_msg}"
);
}
#[tokio::test]
async fn validates_request_hints() {
let trace = LlmTrace::single_turn(
"test-model",
"say hello please",
vec![TraceStep {
request_hint: Some(RequestHint {
last_user_message_contains: Some("hello".to_string()),
min_message_count: Some(1),
}),
response: TraceResponse::Text {
content: "matched".to_string(),
input_tokens: 10,
output_tokens: 5,
},
expected_tool_results: Vec::new(),
}],
);
let llm = TraceLlm::from_trace(trace);
let resp = llm
.complete_with_tools(make_request("say hello please"))
.await
.unwrap();
assert_eq!(resp.content.as_deref(), Some("matched"));
assert_eq!(llm.hint_mismatches(), 0);
}
#[tokio::test]
async fn hint_mismatch_warns_but_continues() {
let trace = LlmTrace::single_turn(
"test-model",
"apple",
vec![TraceStep {
request_hint: Some(RequestHint {
last_user_message_contains: Some("banana".to_string()),
min_message_count: Some(5),
}),
response: TraceResponse::Text {
content: "still works".to_string(),
input_tokens: 10,
output_tokens: 5,
},
expected_tool_results: Vec::new(),
}],
);
let llm = TraceLlm::from_trace(trace);
let resp = llm
.complete_with_tools(make_request("apple"))
.await
.unwrap();
assert_eq!(resp.content.as_deref(), Some("still works"));
assert_eq!(llm.hint_mismatches(), 2);
}
#[tokio::test]
async fn from_json_file() {
let fixture_path = concat!(
env!("CARGO_MANIFEST_DIR"),
"/tests/fixtures/llm_traces/simple_text.json"
);
let llm = TraceLlm::from_file(fixture_path).unwrap();
assert_eq!(llm.model_name(), "test-model");
let resp = llm
.complete_with_tools(make_request("anything"))
.await
.unwrap();
assert_eq!(resp.content.as_deref(), Some("Hello from fixture file!"));
assert_eq!(resp.input_tokens, 50);
assert_eq!(resp.output_tokens, 10);
}
#[tokio::test]
async fn complete_text_step() {
let trace = LlmTrace::single_turn("test-model", "hi", vec![text_step("plain text", 30, 8)]);
let llm = TraceLlm::from_trace(trace);
let resp = llm.complete(make_completion_request("hi")).await.unwrap();
assert_eq!(resp.content, "plain text");
assert_eq!(resp.input_tokens, 30);
assert_eq!(resp.output_tokens, 8);
assert_eq!(resp.finish_reason, FinishReason::Stop);
}
#[tokio::test]
async fn complete_errors_on_tool_calls_step() {
let trace = LlmTrace::single_turn(
"test-model",
"hi",
vec![tool_calls_step(vec![simple_tool_call("echo")], 10, 5)],
);
let llm = TraceLlm::from_trace(trace);
let result = llm.complete(make_completion_request("hi")).await;
assert!(result.is_err());
let err_msg = result.unwrap_err().to_string();
assert!(
err_msg.contains("tool_calls"),
"Expected 'tool_calls' in error: {err_msg}"
);
}
#[tokio::test]
async fn captured_requests() {
let trace = LlmTrace::single_turn(
"test-model",
"test",
vec![text_step("resp1", 10, 5), text_step("resp2", 10, 5)],
);
let llm = TraceLlm::from_trace(trace);
llm.complete_with_tools(make_request("first message"))
.await
.unwrap();
llm.complete_with_tools(make_request("second message"))
.await
.unwrap();
let captured = llm.captured_requests();
assert_eq!(captured.len(), 2);
assert_eq!(captured[0].len(), 1);
assert_eq!(captured[0][0].content, "first message");
assert_eq!(captured[1][0].content, "second message");
}
#[test]
fn deserialize_flat_steps_as_single_turn() {
let json = r#"{"model_name": "m", "steps": [
{"response": {"type": "text", "content": "hi", "input_tokens": 1, "output_tokens": 1}}
]}"#;
let trace: LlmTrace = serde_json::from_str(json).unwrap();
assert_eq!(trace.turns.len(), 1);
assert_eq!(trace.turns[0].user_input, "(test input)");
assert_eq!(trace.turns[0].steps.len(), 1);
}
#[test]
fn deserialize_turns_format() {
let json = r#"{"model_name": "m", "turns": [
{"user_input": "hello", "steps": [
{"response": {"type": "text", "content": "hi", "input_tokens": 1, "output_tokens": 1}}
]},
{"user_input": "bye", "steps": [
{"response": {"type": "text", "content": "bye", "input_tokens": 1, "output_tokens": 1}}
]}
]}"#;
let trace: LlmTrace = serde_json::from_str(json).unwrap();
assert_eq!(trace.turns.len(), 2);
assert_eq!(trace.turns[0].user_input, "hello");
assert_eq!(trace.turns[1].user_input, "bye");
}
#[tokio::test]
async fn multi_turn() {
let trace = LlmTrace::new(
"turns-model",
vec![
TraceTurn {
user_input: "first".to_string(),
steps: vec![text_step("turn 1 response", 10, 5)],
expects: TraceExpects::default(),
},
TraceTurn {
user_input: "second".to_string(),
steps: vec![text_step("turn 2 response", 20, 10)],
expects: TraceExpects::default(),
},
],
);
let llm = TraceLlm::from_trace(trace);
let resp1 = llm
.complete_with_tools(make_request("first"))
.await
.unwrap();
assert_eq!(resp1.content.as_deref(), Some("turn 1 response"));
let resp2 = llm
.complete_with_tools(make_request("second"))
.await
.unwrap();
assert_eq!(resp2.content.as_deref(), Some("turn 2 response"));
assert_eq!(llm.calls(), 2);
}
}
// ---------------------------------------------------------------------------
// test_rig
// ---------------------------------------------------------------------------
#[cfg(feature = "libsql")]
mod test_rig_tests {
use std::time::Duration;
use crate::support::test_rig::TestRigBuilder;
use crate::support::trace_llm::{LlmTrace, TraceResponse, TraceStep};
#[tokio::test]
async fn rig_builds_and_runs() {
let trace = LlmTrace::single_turn(
"test-model",
"Hello test rig",
vec![TraceStep {
request_hint: None,
response: TraceResponse::Text {
content: "I am the test rig response.".to_string(),
input_tokens: 50,
output_tokens: 15,
},
expected_tool_results: Vec::new(),
}],
);
let rig = TestRigBuilder::new().with_trace(trace).build().await;
rig.send_message("Hello test rig").await;
let responses = rig.wait_for_responses(1, Duration::from_secs(10)).await;
assert!(
!responses.is_empty(),
"Expected at least one response from the agent"
);
let found = responses
.iter()
.any(|r| r.content.contains("I am the test rig response."));
assert!(
found,
"Expected a response containing the trace text, got: {:?}",
responses.iter().map(|r| &r.content).collect::<Vec<_>>()
);
rig.shutdown();
}
}