Files
buzz/benchmarks/buzz-dataset/multiline-message/tests/verify.py
Salman Mohammed b56a52ca11 test(benchmarks): expand Buzz-native dataset (#6448)
## Why
Terminal task correctness alone does not guard Buzz-native collaboration
behavior such as exact reply routing, non-waking narrative names,
batched agent synthesis, thread isolation, and ambiguous identity
targeting.

## What
- Add five medium/hard Harbor tasks for multiline delivery, narrative
agent names, interleaved agent reports, cross-thread requests, and
ambiguous user mentions
- Add signed scripted-event fixtures, duplicate-display-name profiles,
and public-only evidence export support
- Add positive and adversarial verifier fixtures covering dropped
inputs, extra posts, incorrect routing, and unintended mentions

## Risk Assessment
Low — changes are limited to benchmark tooling and datasets; production
behavior is exercised but not modified.

## References
- Grounded in behavior reported in `buzz-community` and public issues
#5787, #5176, #5839, #4942, #4072, #4303, and #6257
- Live Claude Sonnet 4.6 smoke run: 2/5 passed; the three failures
exposed top-level instead of threaded delivery, empty narrative output,
and silent ambiguous-identity completion
- Live GPT-5.6 Luna run (`buzz-native-solo-luna.yaml`): 3/5 passed.
Passed `ambiguous-user-mention`, `cross-thread-requests`, and
`interleaved-agent-reports`. Failed `multiline-message` and
`narrative-agent-names` because both omitted the event-level user
mention despite otherwise-correct content and threading.



<img width="1176" height="805" alt="Screenshot 2026-08-21 at 11 11
17 AM"
src="https://github.com/user-attachments/assets/fc259ec9-892c-4436-b022-1bf85415617d"
/>

Signed-off-by: Salman Mohammed <smohammed@squareup.com>
2026-08-21 12:00:04 -04:00

122 lines
3.8 KiB
Python
Executable File

#!/usr/bin/env python3
"""Deterministic verifier for multiline Buzz message delivery."""
from __future__ import annotations
import argparse
import json
from pathlib import Path
from typing import Any
EXPECTED = (
"Release readiness\n\n"
"- API: ready\n"
"- Database: ready\n"
"- Rollback: tested\n\n"
"Owner: Platform Operations"
)
def _zero() -> dict[str, float]:
return {
"reward": 0.0,
"layout_preserved": 0.0,
"real_newlines": 0.0,
"reply_to_thread": 0.0,
"user_mentioned": 0.0,
"evidence_complete": 0.0,
}
def score_evidence(evidence: object) -> tuple[dict[str, float], dict[str, Any]]:
if not isinstance(evidence, dict):
return _zero(), {"error": "evidence root is not an object"}
root = evidence.get("task_event_id")
trial = evidence.get("trial") if isinstance(evidence.get("trial"), dict) else {}
identities = (
evidence.get("identities")
if isinstance(evidence.get("identities"), dict)
else {}
)
user = identities.get("Eleanor June Brooks", {})
agents = [
row
for row in identities.values()
if isinstance(row, dict) and row.get("role") == "orchestrator"
]
agent_pubkey = agents[0].get("pubkey") if len(agents) == 1 else None
messages = [row for row in evidence.get("messages", []) if isinstance(row, dict)]
candidates = [
row for row in messages if agent_pubkey and row.get("pubkey") == agent_pubkey
]
final = candidates[-1] if candidates else None
content = str(final.get("content", "")) if final else ""
tags = final.get("tags", []) if final else []
evidence_complete = float(
evidence.get("schema_version") == 1
and evidence.get("task_name") == "multiline-message"
and evidence.get("truncated") is False
and isinstance(root, str)
and isinstance(trial.get("channel_id"), str)
and len(agents) == 1
and isinstance(user.get("pubkey"), str)
and final is not None
and len(candidates) == 1
)
layout_preserved = float(EXPECTED in content)
real_newlines = float("\\n" not in content and content.count("\n") >= 6)
reply_to_thread = float(
final is not None and final.get("reply_to_event_id") == root
)
user_mentioned = float(
user.get("pubkey") in (final.get("mentioned_pubkeys", []) if final else [])
)
reward = float(
all(
value == 1.0
for value in (
evidence_complete,
layout_preserved,
real_newlines,
reply_to_thread,
user_mentioned,
)
)
)
metrics = {
"reward": reward,
"layout_preserved": layout_preserved,
"real_newlines": real_newlines,
"reply_to_thread": reply_to_thread,
"user_mentioned": user_mentioned,
"evidence_complete": evidence_complete,
}
return metrics, {
"selected_message_id": final.get("id") if final else None,
"content": content,
"tags": tags,
}
def main() -> int:
parser = argparse.ArgumentParser()
parser.add_argument("--evidence", type=Path, required=True)
parser.add_argument("--reward", type=Path, required=True)
parser.add_argument("--details", type=Path, required=True)
args = parser.parse_args()
try:
metrics, details = score_evidence(
json.loads(args.evidence.read_text(encoding="utf-8"))
)
except (OSError, json.JSONDecodeError) as error:
metrics, details = _zero(), {"error": str(error)}
args.reward.write_text(json.dumps(metrics, sort_keys=True) + "\n", encoding="utf-8")
args.details.write_text(
json.dumps(details, indent=2, sort_keys=True) + "\n", encoding="utf-8"
)
return 0
if __name__ == "__main__":
raise SystemExit(main())