925fff731f
- Memorizer tracks user_expectation (conversational/delegated/waiting_input/observing) - Output node adjusts phrasing per expectation - PA retry loop: reformulates job on expert failure (all retries exhausted or tool skip) - Machine state in PA context: get_machine_summary includes current state, buttons, stored data - Expert writes to machine state via update_machine + transition_machine - Expanded baked schema coverage - Awareness panel shows color-coded expectation state - Dashboard and workspace component updates Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
1033 lines
34 KiB
JSON
1033 lines
34 KiB
JSON
{
|
|
"timestamp": "2026-03-29 12:46:11",
|
|
"testcases": {
|
|
"Button Persistence": [
|
|
{
|
|
"step": "Setup",
|
|
"check": "clear",
|
|
"status": "PASS",
|
|
"detail": "cleared"
|
|
},
|
|
{
|
|
"step": "Create buttons",
|
|
"check": "send: create two buttons: Poodle Bark and Bolo",
|
|
"status": "PASS",
|
|
"detail": "response: Okay, two buttons have been created: Poodle Bark and Bolonka Bark. Click either "
|
|
},
|
|
{
|
|
"step": "Create buttons",
|
|
"check": "actions: length >= 2",
|
|
"status": "PASS",
|
|
"detail": "2 actions >= 2"
|
|
},
|
|
{
|
|
"step": "Create buttons",
|
|
"check": "actions: any action contains \"poodle\" or \"Poodle\"",
|
|
"status": "PASS",
|
|
"detail": "found 'poodle' in actions"
|
|
},
|
|
{
|
|
"step": "Create buttons",
|
|
"check": "actions: any action contains \"bolonka\" or \"Bolonka\"",
|
|
"status": "PASS",
|
|
"detail": "found 'bolonka' in actions"
|
|
},
|
|
{
|
|
"step": "Ask unrelated question (buttons must survive)",
|
|
"check": "send: what time is it?",
|
|
"status": "PASS",
|
|
"detail": "response: It is 12:37 PM.\n"
|
|
},
|
|
{
|
|
"step": "Ask unrelated question (buttons must survive)",
|
|
"check": "response: contains \":\" or \"time\" or \"clock\"",
|
|
"status": "PASS",
|
|
"detail": "found ':'"
|
|
},
|
|
{
|
|
"step": "Ask unrelated question (buttons must survive)",
|
|
"check": "actions: any action contains \"poodle\" or \"Poodle\"",
|
|
"status": "PASS",
|
|
"detail": "found 'poodle' in actions"
|
|
},
|
|
{
|
|
"step": "Ask unrelated question (buttons must survive)",
|
|
"check": "actions: any action contains \"bolonka\" or \"Bolonka\"",
|
|
"status": "PASS",
|
|
"detail": "found 'bolonka' in actions"
|
|
},
|
|
{
|
|
"step": "Ask another question (buttons still there)",
|
|
"check": "send: say hello in German",
|
|
"status": "PASS",
|
|
"detail": "response: Hallo!\n"
|
|
},
|
|
{
|
|
"step": "Ask another question (buttons still there)",
|
|
"check": "response: contains \"Hallo\" or \"hallo\" or \"German\"",
|
|
"status": "PASS",
|
|
"detail": "found 'Hallo'"
|
|
},
|
|
{
|
|
"step": "Ask another question (buttons still there)",
|
|
"check": "actions: any action contains \"poodle\" or \"Poodle\"",
|
|
"status": "PASS",
|
|
"detail": "found 'poodle' in actions"
|
|
},
|
|
{
|
|
"step": "Explicitly replace buttons",
|
|
"check": "send: remove all buttons and create one button",
|
|
"status": "PASS",
|
|
"detail": "response: All buttons have been removed and a Reset button has been created.\n"
|
|
},
|
|
{
|
|
"step": "Explicitly replace buttons",
|
|
"check": "actions: length >= 1",
|
|
"status": "PASS",
|
|
"detail": "1 actions >= 1"
|
|
},
|
|
{
|
|
"step": "Explicitly replace buttons",
|
|
"check": "actions: any action contains \"reset\" or \"Reset\"",
|
|
"status": "PASS",
|
|
"detail": "found 'reset' in actions"
|
|
}
|
|
],
|
|
"Counter State": [
|
|
{
|
|
"step": "Setup",
|
|
"check": "clear",
|
|
"status": "PASS",
|
|
"detail": "cleared"
|
|
},
|
|
{
|
|
"step": "Create counter",
|
|
"check": "send: create a counter starting at 0 with incr",
|
|
"status": "PASS",
|
|
"detail": "response: Counter is up and running! You've got buttons to increment (+1), decrement (-1),"
|
|
},
|
|
{
|
|
"step": "Create counter",
|
|
"check": "response: contains \"counter\" or \"count\"",
|
|
"status": "PASS",
|
|
"detail": "found 'counter'"
|
|
},
|
|
{
|
|
"step": "Create counter",
|
|
"check": "actions: length >= 2",
|
|
"status": "PASS",
|
|
"detail": "3 actions >= 2"
|
|
},
|
|
{
|
|
"step": "Create counter",
|
|
"check": "actions: any action contains \"increment\" or \"inc\" or \"plus\" or \"add\"",
|
|
"status": "PASS",
|
|
"detail": "found 'increment' in actions"
|
|
},
|
|
{
|
|
"step": "Create counter",
|
|
"check": "actions: any action contains \"decrement\" or \"dec\" or \"minus\" or \"sub\"",
|
|
"status": "PASS",
|
|
"detail": "found 'decrement' in actions"
|
|
},
|
|
{
|
|
"step": "Check state",
|
|
"check": "state: topic contains \"counter\" or \"count\" or \"button\"",
|
|
"status": "PASS",
|
|
"detail": "topic=Navigation Machine: creating a counter and buttons contains 'counter'"
|
|
},
|
|
{
|
|
"step": "Ask for current value",
|
|
"check": "send: what is the current count?",
|
|
"status": "PASS",
|
|
"detail": "response: The current count is 0.\n"
|
|
},
|
|
{
|
|
"step": "Ask for current value",
|
|
"check": "response: contains \"0\" or \"zero\"",
|
|
"status": "PASS",
|
|
"detail": "found '0'"
|
|
},
|
|
{
|
|
"step": "Increment",
|
|
"check": "action: machine_action",
|
|
"status": "PASS",
|
|
"detail": "response: The counter is now at 2.\n"
|
|
},
|
|
{
|
|
"step": "Increment",
|
|
"check": "response: contains \"1\" or \"one\" or \"increment\" or \"Navigated\"",
|
|
"status": "FAIL",
|
|
"detail": "none of ['1', 'one', 'increment', 'Navigated'] found in: The counter is now at 2.\n"
|
|
},
|
|
{
|
|
"step": "Increment again",
|
|
"check": "action: machine_action",
|
|
"status": "PASS",
|
|
"detail": "response: The counter is now at 3.\n"
|
|
},
|
|
{
|
|
"step": "Increment again",
|
|
"check": "response: contains \"2\" or \"two\" or \"increment\" or \"Navigated\"",
|
|
"status": "FAIL",
|
|
"detail": "none of ['2', 'two', 'increment', 'Navigated'] found in: The counter is now at 3.\n"
|
|
},
|
|
{
|
|
"step": "Decrement",
|
|
"check": "action: machine_action",
|
|
"status": "PASS",
|
|
"detail": "response: The counter is now at 5.\n"
|
|
},
|
|
{
|
|
"step": "Decrement",
|
|
"check": "response: contains \"1\" or \"one\" or \"decrement\" or \"Navigated\"",
|
|
"status": "FAIL",
|
|
"detail": "none of ['1', 'one', 'decrement', 'Navigated'] found in: The counter is now at 5.\n"
|
|
},
|
|
{
|
|
"step": "Verify memorizer tracks it",
|
|
"check": "state: topic contains \"count\"",
|
|
"status": "PASS",
|
|
"detail": "topic=Navigation Machine: reading the counter contains 'count'"
|
|
}
|
|
],
|
|
"DB Exploration": [
|
|
{
|
|
"step": "Setup",
|
|
"check": "clear",
|
|
"status": "PASS",
|
|
"detail": "cleared"
|
|
},
|
|
{
|
|
"step": "Query renders table in workspace",
|
|
"check": "send: show me 5 customers from the database",
|
|
"status": "PASS",
|
|
"detail": "response: OK, I found 5 customers in the database: Kathrin Jager, Leon Schreiner, Ralf Rit"
|
|
},
|
|
{
|
|
"step": "Query renders table in workspace",
|
|
"check": "trace: has tool_call",
|
|
"status": "PASS",
|
|
"detail": "found event 'tool_call'"
|
|
},
|
|
{
|
|
"step": "Query renders table in workspace",
|
|
"check": "actions: has table",
|
|
"status": "PASS",
|
|
"detail": "table found: 23 cols, 5 rows"
|
|
},
|
|
{
|
|
"step": "Query renders table in workspace",
|
|
"check": "response: not contains \"---|\" or \"| ID\"",
|
|
"status": "PASS",
|
|
"detail": "none of ['---|', '| ID'] found (as expected)"
|
|
},
|
|
{
|
|
"step": "Chat summarizes, does not dump data",
|
|
"check": "response: contains \"customer\" or \"Kunde\" or \"5\" or \"table\"",
|
|
"status": "PASS",
|
|
"detail": "found 'customer'"
|
|
},
|
|
{
|
|
"step": "Chat summarizes, does not dump data",
|
|
"check": "response: length > 10",
|
|
"status": "PASS",
|
|
"detail": "length 128 > 10"
|
|
},
|
|
{
|
|
"step": "Thinker builds exploration UI (not describes it)",
|
|
"check": "send: select customer 2 Kathrin Jager, add but",
|
|
"status": "PASS",
|
|
"detail": "response: "
|
|
},
|
|
{
|
|
"step": "Thinker builds exploration UI (not describes it)",
|
|
"check": "actions: length >= 1",
|
|
"status": "PASS",
|
|
"detail": "2 actions >= 1"
|
|
},
|
|
{
|
|
"step": "Thinker builds exploration UI (not describes it)",
|
|
"check": "response: not contains \"UI team\" or \"will add\" or \"will create\"",
|
|
"status": "PASS",
|
|
"detail": "none of ['UI team', 'will add', 'will create'] found (as expected)"
|
|
},
|
|
{
|
|
"step": "Error recovery on bad query",
|
|
"check": "send: SELECT * FROM nichtexistiert LIMIT 5",
|
|
"status": "PASS",
|
|
"detail": "response: I can't explore the objects and devices for Kathrin Jager right now. I tried to "
|
|
},
|
|
{
|
|
"step": "Error recovery on bad query",
|
|
"check": "trace: has tool_call",
|
|
"status": "PASS",
|
|
"detail": "found event 'tool_call'"
|
|
},
|
|
{
|
|
"step": "Error recovery on bad query",
|
|
"check": "response: not contains \"1146\"",
|
|
"status": "PASS",
|
|
"detail": "none of ['1146'] found (as expected)"
|
|
},
|
|
{
|
|
"step": "Error recovery on bad query",
|
|
"check": "response: length > 10",
|
|
"status": "PASS",
|
|
"detail": "length 226 > 10"
|
|
}
|
|
],
|
|
"Director Node": [
|
|
{
|
|
"step": "Setup",
|
|
"check": "clear",
|
|
"status": "PASS",
|
|
"detail": "cleared"
|
|
},
|
|
{
|
|
"step": "Casual chat establishes mode",
|
|
"check": "send: hey, just hanging out, what's up?",
|
|
"status": "PASS",
|
|
"detail": "response: Just checking in! It looks like you're having a good time. Want to dive into the"
|
|
},
|
|
{
|
|
"step": "Casual chat establishes mode",
|
|
"check": "response: length > 5",
|
|
"status": "PASS",
|
|
"detail": "length 128 > 5"
|
|
},
|
|
{
|
|
"step": "Casual chat establishes mode",
|
|
"check": "trace: has director_updated",
|
|
"status": "FAIL",
|
|
"detail": "no 'director_updated' event in trace"
|
|
},
|
|
{
|
|
"step": "Director picks up frustration",
|
|
"check": "send: ugh this is so annoying, nothing makes s",
|
|
"status": "PASS",
|
|
"detail": "response: I hear you. Database schemas are messy. Let's slow down: what exactly are you tr"
|
|
},
|
|
{
|
|
"step": "Director picks up frustration",
|
|
"check": "response: length > 10",
|
|
"status": "PASS",
|
|
"detail": "length 200 > 10"
|
|
},
|
|
{
|
|
"step": "Director picks up frustration",
|
|
"check": "trace: has director_updated",
|
|
"status": "FAIL",
|
|
"detail": "no 'director_updated' event in trace"
|
|
},
|
|
{
|
|
"step": "Switch to building mode",
|
|
"check": "send: ok let's build a todo list app",
|
|
"status": "PASS",
|
|
"detail": "response: "
|
|
},
|
|
{
|
|
"step": "Switch to building mode",
|
|
"check": "response: length > 10",
|
|
"status": "FAIL",
|
|
"detail": "length 0 <= 10"
|
|
},
|
|
{
|
|
"step": "Switch to building mode",
|
|
"check": "trace: has director_updated",
|
|
"status": "FAIL",
|
|
"detail": "no 'director_updated' event in trace"
|
|
}
|
|
],
|
|
"Pub Conversation": [
|
|
{
|
|
"step": "Setup",
|
|
"check": "clear",
|
|
"status": "PASS",
|
|
"detail": "cleared"
|
|
},
|
|
{
|
|
"step": "Set the scene",
|
|
"check": "send: Hey, Alice and I are heading to the pub ",
|
|
"status": "PASS",
|
|
"detail": "response: That sounds great! Have fun tonight with Alice at the pub. 🍺 Let me know if anyt"
|
|
},
|
|
{
|
|
"step": "Set the scene",
|
|
"check": "response: length > 10",
|
|
"status": "PASS",
|
|
"detail": "length 147 > 10"
|
|
},
|
|
{
|
|
"step": "Set the scene",
|
|
"check": "state: situation contains \"pub\" or \"Alice\"",
|
|
"status": "FAIL",
|
|
"detail": "situation=authenticated on https://cog.loop42.de doesn't contain any of ['pub', 'Alice']"
|
|
},
|
|
{
|
|
"step": "Language switch to German",
|
|
"check": "send: Wir sind jetzt im Biergarten angekommen",
|
|
"status": "PASS",
|
|
"detail": "response: Ah, wunderbar that you've arrived safe and sound! Hope you enjoy the Biergarten "
|
|
},
|
|
{
|
|
"step": "Language switch to German",
|
|
"check": "response: length > 10",
|
|
"status": "PASS",
|
|
"detail": "length 101 > 10"
|
|
},
|
|
{
|
|
"step": "Language switch to German",
|
|
"check": "state: language is \"de\" or \"mixed\"",
|
|
"status": "PASS",
|
|
"detail": "language=mixed"
|
|
},
|
|
{
|
|
"step": "Context awareness",
|
|
"check": "send: Was sollen wir bestellen?",
|
|
"status": "PASS",
|
|
"detail": "response: Da Alice schon ein Hefeweizen bestellt hat, wie wäre es mit einem erfrischenden "
|
|
},
|
|
{
|
|
"step": "Context awareness",
|
|
"check": "response: length > 10",
|
|
"status": "PASS",
|
|
"detail": "length 211 > 10"
|
|
},
|
|
{
|
|
"step": "Context awareness",
|
|
"check": "state: topic contains \"bestell\" or \"order\" or \"pub\" or \"Biergarten\"",
|
|
"status": "PASS",
|
|
"detail": "topic=Biergarten visit contains 'Biergarten'"
|
|
},
|
|
{
|
|
"step": "Alice speaks",
|
|
"check": "send: Alice says: I'll have a Hefeweizen pleas",
|
|
"status": "PASS",
|
|
"detail": "response: Okay, one Hefeweizen coming right up for Alice! 🍻\n"
|
|
},
|
|
{
|
|
"step": "Alice speaks",
|
|
"check": "response: length > 10",
|
|
"status": "PASS",
|
|
"detail": "length 50 > 10"
|
|
},
|
|
{
|
|
"step": "Alice speaks",
|
|
"check": "state: facts any contains \"Alice\" or \"Hefeweizen\"",
|
|
"status": "PASS",
|
|
"detail": "found 'Alice' in facts"
|
|
},
|
|
{
|
|
"step": "Ask for time (tool use)",
|
|
"check": "send: wie spaet ist es eigentlich?",
|
|
"status": "PASS",
|
|
"detail": "response: Aktuell ist es 12:44 Uhr. Die Zeit fliegt im Biergarten! 😉\n"
|
|
},
|
|
{
|
|
"step": "Ask for time (tool use)",
|
|
"check": "response: matches \\d{1,2}:\\d{2}",
|
|
"status": "PASS",
|
|
"detail": "matched /\\d{1,2}:\\d{2}/"
|
|
},
|
|
{
|
|
"step": "Back to English",
|
|
"check": "send: Let's switch to English, what was the la",
|
|
"status": "PASS",
|
|
"detail": "response: Alice said, \"I'll have a Hefeweizen please.\"\n"
|
|
},
|
|
{
|
|
"step": "Back to English",
|
|
"check": "state: language is \"en\" or \"mixed\"",
|
|
"status": "PASS",
|
|
"detail": "language=mixed"
|
|
},
|
|
{
|
|
"step": "Back to English",
|
|
"check": "response: contains \"Alice\" or \"Hefeweizen\"",
|
|
"status": "PASS",
|
|
"detail": "found 'Alice'"
|
|
},
|
|
{
|
|
"step": "Mood check",
|
|
"check": "send: This is really fun!",
|
|
"status": "PASS",
|
|
"detail": "response: Glad you're having a blast! Keep the fun going with Alice—sounds like a perfect "
|
|
},
|
|
{
|
|
"step": "Mood check",
|
|
"check": "state: user_mood is \"happy\" or \"playful\" or \"excited\"",
|
|
"status": "PASS",
|
|
"detail": "user_mood=happy"
|
|
}
|
|
],
|
|
"Reflex Path": [
|
|
{
|
|
"step": "Setup",
|
|
"check": "clear",
|
|
"status": "PASS",
|
|
"detail": "cleared"
|
|
},
|
|
{
|
|
"step": "Greeting triggers reflex",
|
|
"check": "send: hey!",
|
|
"status": "PASS",
|
|
"detail": "response: Servus Nico! 👋 Was geht? 🍻\n"
|
|
},
|
|
{
|
|
"step": "Greeting triggers reflex",
|
|
"check": "response: length > 2",
|
|
"status": "PASS",
|
|
"detail": "length 27 > 2"
|
|
},
|
|
{
|
|
"step": "Greeting triggers reflex",
|
|
"check": "trace: has reflex_path",
|
|
"status": "PASS",
|
|
"detail": "found event 'reflex_path'"
|
|
},
|
|
{
|
|
"step": "Thanks triggers reflex",
|
|
"check": "send: thanks",
|
|
"status": "PASS",
|
|
"detail": "response: Kein Problem! 😊\n"
|
|
},
|
|
{
|
|
"step": "Thanks triggers reflex",
|
|
"check": "response: length > 2",
|
|
"status": "PASS",
|
|
"detail": "length 16 > 2"
|
|
},
|
|
{
|
|
"step": "Thanks triggers reflex",
|
|
"check": "trace: has reflex_path",
|
|
"status": "PASS",
|
|
"detail": "found event 'reflex_path'"
|
|
},
|
|
{
|
|
"step": "Complex request does NOT trigger reflex",
|
|
"check": "send: explain how neural networks work in deta",
|
|
"status": "PASS",
|
|
"detail": "response: Think of your brain: neurons firing, passing signals! Neural networks mimic this"
|
|
},
|
|
{
|
|
"step": "Complex request does NOT trigger reflex",
|
|
"check": "response: length > 20",
|
|
"status": "PASS",
|
|
"detail": "length 759 > 20"
|
|
},
|
|
{
|
|
"step": "Complex request does NOT trigger reflex",
|
|
"check": "trace: input.analysis.intent is \"question\" or \"request\"",
|
|
"status": "PASS",
|
|
"detail": "input.analysis.intent=question"
|
|
},
|
|
{
|
|
"step": "Complex request does NOT trigger reflex",
|
|
"check": "trace: has decided",
|
|
"status": "PASS",
|
|
"detail": "found event 'decided'"
|
|
}
|
|
],
|
|
"S3* Audit Corrections": [
|
|
{
|
|
"step": "Setup",
|
|
"check": "clear",
|
|
"status": "PASS",
|
|
"detail": "cleared"
|
|
},
|
|
{
|
|
"step": "Tool calls produce results (baseline)",
|
|
"check": "send: create two buttons: Alpha and Beta",
|
|
"status": "PASS",
|
|
"detail": "response: Done! Alpha and Beta buttons are live. Hit 'em up. 🎯\n"
|
|
},
|
|
{
|
|
"step": "Tool calls produce results (baseline)",
|
|
"check": "actions: length >= 1",
|
|
"status": "PASS",
|
|
"detail": "2 actions >= 1"
|
|
},
|
|
{
|
|
"step": "Tool calls produce results (baseline)",
|
|
"check": "actions: any action contains \"alpha\" or \"Alpha\"",
|
|
"status": "PASS",
|
|
"detail": "found 'alpha' in actions"
|
|
},
|
|
{
|
|
"step": "Dashboard mismatch triggers re-emit",
|
|
"check": "send: I see nothing on my dashboard, fix it",
|
|
"status": "PASS",
|
|
"detail": "response: Got it—buttons were stuck server-side. Pushing them now. You should see Alpha an"
|
|
},
|
|
{
|
|
"step": "Dashboard mismatch triggers re-emit",
|
|
"check": "response: not contains \"sorry\" or \"apologize\"",
|
|
"status": "PASS",
|
|
"detail": "none of ['sorry', 'apologize'] found (as expected)"
|
|
},
|
|
{
|
|
"step": "Dashboard mismatch triggers re-emit",
|
|
"check": "actions: length >= 1",
|
|
"status": "PASS",
|
|
"detail": "2 actions >= 1"
|
|
},
|
|
{
|
|
"step": "DB error triggers retry with corrected SQL",
|
|
"check": "send: SELECT * FROM NichtExistent LIMIT 5",
|
|
"status": "PASS",
|
|
"detail": "response: Ah, the database is telling me that the table `NichtExistent` doesn't exist. 😉 D"
|
|
},
|
|
{
|
|
"step": "DB error triggers retry with corrected SQL",
|
|
"check": "trace: has tool_call",
|
|
"status": "PASS",
|
|
"detail": "found event 'tool_call'"
|
|
},
|
|
{
|
|
"step": "DB error triggers retry with corrected SQL",
|
|
"check": "response: not contains \"1146\"",
|
|
"status": "PASS",
|
|
"detail": "none of ['1146'] found (as expected)"
|
|
},
|
|
{
|
|
"step": "DB error triggers retry with corrected SQL",
|
|
"check": "response: length > 10",
|
|
"status": "PASS",
|
|
"detail": "length 130 > 10"
|
|
},
|
|
{
|
|
"step": "Complex request gets Director plan",
|
|
"check": "send: investigate which customers have the mos",
|
|
"status": "PASS",
|
|
"detail": "response: ```text\nError: (1054, \"Unknown column 'k.kundenname' in 'SELECT'\")\n```\n\nI'm gett"
|
|
},
|
|
{
|
|
"step": "Complex request gets Director plan",
|
|
"check": "trace: has director_plan or decided",
|
|
"status": "PASS",
|
|
"detail": "found event 'decided'"
|
|
},
|
|
{
|
|
"step": "Complex request gets Director plan",
|
|
"check": "trace: has tool_call",
|
|
"status": "PASS",
|
|
"detail": "found event 'tool_call'"
|
|
},
|
|
{
|
|
"step": "Complex request gets Director plan",
|
|
"check": "response: length > 20",
|
|
"status": "PASS",
|
|
"detail": "length 342 > 20"
|
|
}
|
|
],
|
|
"State Machines": [
|
|
{
|
|
"step": "Setup",
|
|
"check": "clear",
|
|
"status": "PASS",
|
|
"detail": "cleared"
|
|
},
|
|
{
|
|
"step": "Create a machine",
|
|
"check": "send: create a navigation machine called \"nav\"",
|
|
"status": "PASS",
|
|
"detail": "response: Okay, a navigation machine named `nav` has been set up. It starts at the `main`"
|
|
},
|
|
{
|
|
"step": "Create a machine",
|
|
"check": "trace: has tool_call create_machine",
|
|
"status": "PASS",
|
|
"detail": "found create_machine via machine_created event"
|
|
},
|
|
{
|
|
"step": "Create a machine",
|
|
"check": "trace: machine_created id=\"nav\"",
|
|
"status": "PASS",
|
|
"detail": "machine 'nav' created"
|
|
},
|
|
{
|
|
"step": "Verify machine renders",
|
|
"check": "send: what machines are on my dashboard?",
|
|
"status": "PASS",
|
|
"detail": "response: Currently, the dashboard has one machine: `nav` in the `main` state. It's showin"
|
|
},
|
|
{
|
|
"step": "Verify machine renders",
|
|
"check": "response: contains \"nav\" or \"machine\"",
|
|
"status": "PASS",
|
|
"detail": "found 'nav'"
|
|
},
|
|
{
|
|
"step": "Navigate via button click (local transition)",
|
|
"check": "action matching 'menu_1'",
|
|
"status": "FAIL",
|
|
"detail": "no action matching 'menu_1' in [None, None, None]"
|
|
},
|
|
{
|
|
"step": "Navigate via button click (local transition)",
|
|
"check": "trace: has machine_transition",
|
|
"status": "FAIL",
|
|
"detail": "no 'machine_transition' event in trace"
|
|
},
|
|
{
|
|
"step": "Navigate via button click (local transition)",
|
|
"check": "trace: no thinker",
|
|
"status": "PASS",
|
|
"detail": "no 'thinker' event (as expected)"
|
|
},
|
|
{
|
|
"step": "Add a state to existing machine",
|
|
"check": "send: add a state \"sub3\" to the nav machine wi",
|
|
"status": "PASS",
|
|
"detail": "response: State `sub3` has been added to the `nav` machine. It displays \"Third submenu\" an"
|
|
},
|
|
{
|
|
"step": "Add a state to existing machine",
|
|
"check": "trace: has tool_call add_state",
|
|
"status": "FAIL",
|
|
"detail": "no tool_call 'add_state' in trace"
|
|
},
|
|
{
|
|
"step": "Reset machine",
|
|
"check": "send: reset the nav machine to its initial sta",
|
|
"status": "PASS",
|
|
"detail": "response: The `nav` machine has been reset to its initial `main` state.\n"
|
|
},
|
|
{
|
|
"step": "Reset machine",
|
|
"check": "trace: has tool_call reset_machine",
|
|
"status": "PASS",
|
|
"detail": "found reset_machine via machine_reset event"
|
|
},
|
|
{
|
|
"step": "Reset machine",
|
|
"check": "response: contains \"main\" or \"reset\" or \"initial\"",
|
|
"status": "PASS",
|
|
"detail": "found 'main'"
|
|
},
|
|
{
|
|
"step": "Create second machine alongside first",
|
|
"check": "send: create a counter machine called \"clicks\"",
|
|
"status": "PASS",
|
|
"detail": "response: A new state machine called `clicks` has been created, starting at the `zero` sta"
|
|
},
|
|
{
|
|
"step": "Create second machine alongside first",
|
|
"check": "trace: has tool_call create_machine",
|
|
"status": "PASS",
|
|
"detail": "found create_machine via machine_created event"
|
|
},
|
|
{
|
|
"step": "Create second machine alongside first",
|
|
"check": "trace: machine_created id=\"clicks\"",
|
|
"status": "PASS",
|
|
"detail": "machine 'clicks' created"
|
|
},
|
|
{
|
|
"step": "Both machines coexist",
|
|
"check": "send: what machines are running?",
|
|
"status": "PASS",
|
|
"detail": "response: There are two machines running: `nav` (currently in `main`) and `clicks` (curren"
|
|
},
|
|
{
|
|
"step": "Both machines coexist",
|
|
"check": "response: contains \"nav\"",
|
|
"status": "PASS",
|
|
"detail": "found 'nav'"
|
|
},
|
|
{
|
|
"step": "Both machines coexist",
|
|
"check": "response: contains \"click\"",
|
|
"status": "PASS",
|
|
"detail": "found 'click'"
|
|
},
|
|
{
|
|
"step": "Destroy one machine",
|
|
"check": "send: destroy the clicks machine",
|
|
"status": "PASS",
|
|
"detail": "response: The `clicks` machine has been destroyed.\n"
|
|
},
|
|
{
|
|
"step": "Destroy one machine",
|
|
"check": "trace: has tool_call destroy_machine",
|
|
"status": "PASS",
|
|
"detail": "found destroy_machine via machine_destroyed event"
|
|
},
|
|
{
|
|
"step": "Destroy one machine",
|
|
"check": "send: what machines are running?",
|
|
"status": "PASS",
|
|
"detail": "response: Only the `nav` machine is currently running, and it's in the `main` state.\n"
|
|
},
|
|
{
|
|
"step": "Destroy one machine",
|
|
"check": "response: contains \"nav\"",
|
|
"status": "PASS",
|
|
"detail": "found 'nav'"
|
|
}
|
|
],
|
|
"Structured Input Analysis": [
|
|
{
|
|
"step": "Setup",
|
|
"check": "clear",
|
|
"status": "PASS",
|
|
"detail": "cleared"
|
|
},
|
|
{
|
|
"step": "Social greeting",
|
|
"check": "send: hi there!",
|
|
"status": "PASS",
|
|
"detail": "response: Hello Nico! 👋 Anything I can help you with regarding the running machines on the"
|
|
},
|
|
{
|
|
"step": "Social greeting",
|
|
"check": "response: length > 3",
|
|
"status": "PASS",
|
|
"detail": "length 92 > 3"
|
|
},
|
|
{
|
|
"step": "Social greeting",
|
|
"check": "trace: input.analysis.intent is \"social\"",
|
|
"status": "PASS",
|
|
"detail": "input.analysis.intent=social"
|
|
},
|
|
{
|
|
"step": "Social greeting",
|
|
"check": "trace: input.analysis.complexity is \"trivial\"",
|
|
"status": "PASS",
|
|
"detail": "input.analysis.complexity=trivial"
|
|
},
|
|
{
|
|
"step": "Simple request",
|
|
"check": "send: create a counter starting at 0",
|
|
"status": "PASS",
|
|
"detail": "response: Alright, I've created two counter machines for you, both starting at 0. 🚀\n"
|
|
},
|
|
{
|
|
"step": "Simple request",
|
|
"check": "response: length > 10",
|
|
"status": "PASS",
|
|
"detail": "length 74 > 10"
|
|
},
|
|
{
|
|
"step": "Simple request",
|
|
"check": "trace: input.analysis.intent is \"request\" or \"action\"",
|
|
"status": "PASS",
|
|
"detail": "input.analysis.intent=request"
|
|
},
|
|
{
|
|
"step": "Simple request",
|
|
"check": "trace: input.analysis.complexity is \"simple\" or \"complex\"",
|
|
"status": "PASS",
|
|
"detail": "input.analysis.complexity=simple"
|
|
},
|
|
{
|
|
"step": "German question",
|
|
"check": "send: Wie spaet ist es?",
|
|
"status": "PASS",
|
|
"detail": "response: Es ist 12:44 Uhr. ☀️\n"
|
|
},
|
|
{
|
|
"step": "German question",
|
|
"check": "response: length > 5",
|
|
"status": "PASS",
|
|
"detail": "length 21 > 5"
|
|
},
|
|
{
|
|
"step": "German question",
|
|
"check": "trace: input.analysis.language is \"de\"",
|
|
"status": "PASS",
|
|
"detail": "input.analysis.language=de"
|
|
},
|
|
{
|
|
"step": "German question",
|
|
"check": "trace: input.analysis.intent is \"question\"",
|
|
"status": "PASS",
|
|
"detail": "input.analysis.intent=question"
|
|
},
|
|
{
|
|
"step": "Frustrated tone",
|
|
"check": "send: this is broken, nothing works and I'm si",
|
|
"status": "PASS",
|
|
"detail": "response: I hear you, that's frustrating! Let me help figure out what's going wrong. What "
|
|
},
|
|
{
|
|
"step": "Frustrated tone",
|
|
"check": "response: length > 10",
|
|
"status": "PASS",
|
|
"detail": "length 168 > 10"
|
|
},
|
|
{
|
|
"step": "Frustrated tone",
|
|
"check": "trace: input.analysis.tone is \"frustrated\" or \"urgent\"",
|
|
"status": "PASS",
|
|
"detail": "input.analysis.tone=frustrated"
|
|
},
|
|
{
|
|
"step": "Simple acknowledgment",
|
|
"check": "send: ok thanks bye",
|
|
"status": "PASS",
|
|
"detail": "response: Ok, feel free to reach out if you need anything else! 👍\n"
|
|
},
|
|
{
|
|
"step": "Simple acknowledgment",
|
|
"check": "trace: input.analysis.intent is \"social\"",
|
|
"status": "PASS",
|
|
"detail": "input.analysis.intent=social"
|
|
},
|
|
{
|
|
"step": "Simple acknowledgment",
|
|
"check": "trace: input.analysis.complexity is \"trivial\"",
|
|
"status": "PASS",
|
|
"detail": "input.analysis.complexity=trivial"
|
|
}
|
|
],
|
|
"Dashboard Feedback (S3*)": [
|
|
{
|
|
"step": "Setup",
|
|
"check": "clear",
|
|
"status": "PASS",
|
|
"detail": "cleared"
|
|
},
|
|
{
|
|
"step": "Thinker sees buttons in dashboard",
|
|
"check": "send: create two buttons: hello and world",
|
|
"status": "PASS",
|
|
"detail": "response: Done! Two buttons, 'hello' and 'world', have been created for you.\n"
|
|
},
|
|
{
|
|
"step": "Thinker sees buttons in dashboard",
|
|
"check": "actions: length >= 2",
|
|
"status": "PASS",
|
|
"detail": "2 actions >= 2"
|
|
},
|
|
{
|
|
"step": "Thinker sees buttons in dashboard",
|
|
"check": "send: what buttons can you see in my dashboard",
|
|
"status": "PASS",
|
|
"detail": "response: Currently, the dashboard displays two buttons: 'Hello' and 'World'.\n"
|
|
},
|
|
{
|
|
"step": "Thinker sees buttons in dashboard",
|
|
"check": "response: contains \"Hello\" or \"hello\"",
|
|
"status": "PASS",
|
|
"detail": "found 'Hello'"
|
|
},
|
|
{
|
|
"step": "Thinker sees buttons in dashboard",
|
|
"check": "response: contains \"World\" or \"world\"",
|
|
"status": "PASS",
|
|
"detail": "found 'World'"
|
|
},
|
|
{
|
|
"step": "Thinker detects empty dashboard",
|
|
"check": "send: I see nothing in my dashboard, what happ",
|
|
"status": "PASS",
|
|
"detail": "response: Ah, it seems the buttons got stuck during transmission. I've resent them! Could "
|
|
},
|
|
{
|
|
"step": "Thinker detects empty dashboard",
|
|
"check": "response: contains \"button\" or \"fix\" or \"restore\" or \"create\" or \"empty\"",
|
|
"status": "PASS",
|
|
"detail": "found 'button'"
|
|
},
|
|
{
|
|
"step": "Dashboard state flows to thinker context",
|
|
"check": "send: create a counter starting at 5",
|
|
"status": "PASS",
|
|
"detail": "response: Counter created and ready to go! You should now see it on your dashboard startin"
|
|
},
|
|
{
|
|
"step": "Dashboard state flows to thinker context",
|
|
"check": "actions: length >= 1",
|
|
"status": "PASS",
|
|
"detail": "2 actions >= 1"
|
|
},
|
|
{
|
|
"step": "Dashboard state flows to thinker context",
|
|
"check": "send: what does my dashboard show?",
|
|
"status": "PASS",
|
|
"detail": "response: Your dashboard currently shows a counter at 5, along with +1 and -1 buttons to a"
|
|
},
|
|
{
|
|
"step": "Dashboard state flows to thinker context",
|
|
"check": "response: contains \"5\" or \"count\"",
|
|
"status": "PASS",
|
|
"detail": "found '5'"
|
|
}
|
|
],
|
|
"Dashboard Mismatch Recovery": [
|
|
{
|
|
"step": "Setup",
|
|
"check": "clear",
|
|
"status": "PASS",
|
|
"detail": "cleared"
|
|
},
|
|
{
|
|
"step": "Create buttons",
|
|
"check": "send: create two buttons: red and blue",
|
|
"status": "PASS",
|
|
"detail": "response: Okay, two buttons, one red and one blue, are now ready for you.\n"
|
|
},
|
|
{
|
|
"step": "Create buttons",
|
|
"check": "actions: length >= 2",
|
|
"status": "PASS",
|
|
"detail": "3 actions >= 2"
|
|
},
|
|
{
|
|
"step": "Dashboard empty — Thinker re-emits",
|
|
"check": "send: I clicked red but nothing happened",
|
|
"status": "PASS",
|
|
"detail": "response: Ah, it seems the buttons weren't wired up correctly. They should be functional n"
|
|
},
|
|
{
|
|
"step": "Dashboard empty — Thinker re-emits",
|
|
"check": "response: contains \"button\" or \"red\" or \"blue\"",
|
|
"status": "PASS",
|
|
"detail": "found 'button'"
|
|
},
|
|
{
|
|
"step": "Dashboard empty — Thinker re-emits",
|
|
"check": "actions: length >= 1",
|
|
"status": "PASS",
|
|
"detail": "3 actions >= 1"
|
|
},
|
|
{
|
|
"step": "Create counter",
|
|
"check": "send: create a counter starting at 0",
|
|
"status": "PASS",
|
|
"detail": "response: A counter initialized to 0 is now displayed, along with increment and decrement "
|
|
},
|
|
{
|
|
"step": "Create counter",
|
|
"check": "actions: length >= 1",
|
|
"status": "PASS",
|
|
"detail": "3 actions >= 1"
|
|
},
|
|
{
|
|
"step": "Counter missing from dashboard — Thinker recovers",
|
|
"check": "send: the dashboard is broken, I only see old ",
|
|
"status": "PASS",
|
|
"detail": "response: I've detected a discrepancy between what the server sent and what you're seeing."
|
|
},
|
|
{
|
|
"step": "Counter missing from dashboard — Thinker recovers",
|
|
"check": "response: contains \"counter\" or \"count\" or \"fix\" or \"recreat\" or \"refresh\" or \"button\" or \"update\"",
|
|
"status": "FAIL",
|
|
"detail": "none of ['counter', 'count', 'fix', 'recreat', 'refresh', 'button', 'update'] found in: I've detected a discrepancy between what the server sent and what you're seeing. I'm re-sending the "
|
|
},
|
|
{
|
|
"step": "Counter missing from dashboard — Thinker recovers",
|
|
"check": "actions: length >= 1",
|
|
"status": "PASS",
|
|
"detail": "6 actions >= 1"
|
|
}
|
|
]
|
|
},
|
|
"summary": {
|
|
"passed": 155,
|
|
"failed": 12
|
|
}
|
|
} |