steve
4e179259de
The session-tool TYPES already lived in tool/ (P4 move) but the executor never used them. This wires them, unblocking artifact-producing host surfaces (mort's chat API / chatbot / .skill / scaddy) to run on executus: - run/session.go: steerMailbox (thread-safe message queue) + runSession (tool.AgentSession over it: AttachImages → a user-role multimodal message injected before the agent's next step) + runPostRun (panic-isolated hook call). - executor: create the mailbox + set inv.AttachImages BEFORE the toolbox build; add inv.ExtraTools + a SessionToolFactory's per-run Tools to the toolbox; defer its Cleanup; merge the session mailbox with the critic's nudges into ONE WithSteer; after the run, call PostRun with the full transcript (runRes.Messages) → Result.PostRunResult (best-effort, never fails the run). - run.Result += PostRunResult *tool.PostRunResult. - dropped the now-dead criticBinding.steerOptions (superseded by drainSteer). Tests: a factory whose PostRun emits an artifact from the output+transcript + Cleanup lands on Result.PostRunResult; a factory-added tool is callable. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
95 lines
3.1 KiB
Go
95 lines
3.1 KiB
Go
package run_test
|
|
|
|
import (
|
|
"context"
|
|
"testing"
|
|
|
|
"gitea.stevedudenhoeffer.com/steve/majordomo/llm"
|
|
"gitea.stevedudenhoeffer.com/steve/majordomo/provider/fake"
|
|
|
|
"gitea.stevedudenhoeffer.com/steve/executus/run"
|
|
"gitea.stevedudenhoeffer.com/steve/executus/tool"
|
|
)
|
|
|
|
// TestSessionToolFactoryPostRun: a SessionToolFactory's PostRun hook produces an
|
|
// artifact (from the run output + transcript) that lands on Result.PostRunResult,
|
|
// and its Cleanup is deferred.
|
|
func TestSessionToolFactoryPostRun(t *testing.T) {
|
|
fp := fake.New("fake")
|
|
fp.Enqueue("m", fake.Reply("hello artifacts"))
|
|
m, _ := fp.Model("m")
|
|
|
|
cleanupCalled := false
|
|
inv := tool.Invocation{
|
|
RunID: "r1",
|
|
SessionToolFactory: func(_ tool.AgentSession) tool.SessionTools {
|
|
return tool.SessionTools{
|
|
PostRun: func(_ context.Context, transcript []llm.Message, output string, _ error) *tool.PostRunResult {
|
|
return &tool.PostRunResult{
|
|
Artifacts: []tool.Artifact{{Name: "out.txt", MimeType: "text/plain", Data: []byte(output)}},
|
|
Metadata: map[string]any{"transcript_len": len(transcript)},
|
|
}
|
|
},
|
|
Cleanup: func() { cleanupCalled = true },
|
|
}
|
|
},
|
|
}
|
|
ex := run.New(run.Config{
|
|
Registry: tool.NewRegistry(),
|
|
Models: func(ctx context.Context, _ string) (context.Context, llm.Model, error) { return ctx, m, nil },
|
|
})
|
|
res := ex.Run(context.Background(), run.RunnableAgent{ModelTier: "m"}, inv, "go")
|
|
if res.Err != nil {
|
|
t.Fatalf("run error: %v", res.Err)
|
|
}
|
|
if res.PostRunResult == nil {
|
|
t.Fatal("Result.PostRunResult is nil — PostRun hook not invoked / not attached")
|
|
}
|
|
if n := len(res.PostRunResult.Artifacts); n != 1 {
|
|
t.Fatalf("artifacts = %d, want 1", n)
|
|
}
|
|
a := res.PostRunResult.Artifacts[0]
|
|
if a.Name != "out.txt" || string(a.Data) != "hello artifacts" {
|
|
t.Errorf("artifact = {%q, %q}", a.Name, string(a.Data))
|
|
}
|
|
if tl, _ := res.PostRunResult.Metadata["transcript_len"].(int); tl < 1 {
|
|
t.Errorf("transcript not passed to PostRun (len=%d)", tl)
|
|
}
|
|
if !cleanupCalled {
|
|
t.Error("Cleanup was not deferred/called")
|
|
}
|
|
}
|
|
|
|
// TestSessionToolFactoryAddsTool: tools the factory returns join the run's
|
|
// toolbox and are callable by the model.
|
|
func TestSessionToolFactoryAddsTool(t *testing.T) {
|
|
fp := fake.New("fake")
|
|
fp.Enqueue("m",
|
|
fake.ReplyWith(llm.Response{ToolCalls: []llm.ToolCall{{ID: "c1", Name: "render", Arguments: []byte(`{}`)}}}),
|
|
fake.Reply("rendered"),
|
|
)
|
|
m, _ := fp.Model("m")
|
|
|
|
toolCalled := false
|
|
renderTool := llm.DefineTool("render", "render a preview",
|
|
func(_ context.Context, _ struct{}) (any, error) { toolCalled = true; return "ok", nil })
|
|
inv := tool.Invocation{
|
|
RunID: "r2",
|
|
SessionToolFactory: func(_ tool.AgentSession) tool.SessionTools {
|
|
return tool.SessionTools{Tools: []llm.Tool{renderTool}}
|
|
},
|
|
}
|
|
ex := run.New(run.Config{
|
|
Registry: tool.NewRegistry(),
|
|
Models: func(ctx context.Context, _ string) (context.Context, llm.Model, error) { return ctx, m, nil },
|
|
})
|
|
res := ex.Run(context.Background(),
|
|
run.RunnableAgent{ModelTier: "m", MaxIterations: 5}, inv, "go")
|
|
if res.Err != nil {
|
|
t.Fatalf("run error: %v", res.Err)
|
|
}
|
|
if !toolCalled {
|
|
t.Error("session-factory tool was not added to the toolbox / not called")
|
|
}
|
|
}
|