AI agent skill that researches any topic across Reddit, X, YouTube, HN, Polymarket, and the web - then synthesizes a grounded summary
Python ai-promptsai-skillblueskyclaudeclaude-codeclawhubdeep-researchhackernewsinstagramopenclawpolymarketrecencyredditresearchsocial-mediatiktoktrendstwitterweb-searchyoutube
"""Qualifier-only GitHub topics must not embed unbounded raw topics.
Stream fanout calls ``search_github`` once per subquery. Before #954 the
error envelope used ``{topic!r}`` and the strip log used the full core, so
log volume, error-report volume, and doctor-hint size scaled with both
fanout and topic length.
"""
from unittest.mock import patch
from lib import github
_FROM = "2026-07-01"
_TO = "2026-07-31"
# Wrapper text plus a ~120-char topic slice. The pre-fix error embedded
# the entire raw topic (thousands of chars on a pathological planner query).
_ERROR_BOUND = 300
_STRIP_LOG_BOUND = 400
def _long_qualifier_only_topic() -> str:
return "created:>2025-03-20 " + ("stars:>1 " * 400)
def test_qualifier_only_topic_returns_clean_empty_envelope_without_network():
"""#953 supersedes the #954 error bound: a qualifier-only topic is a clean
no-results envelope (no error key), so there is nothing to spam."""
topic = _long_qualifier_only_topic()
with patch.object(github, "_resolve_token", return_value="t"), patch.object(
github, "_fetch_json"
) as fetch:
result = github.search_github(topic, _FROM, _TO)
fetch.assert_not_called()
assert result["items"] == []
assert "error" not in result
def test_short_qualifier_only_topic_is_also_clean_no_results():
topic = "created:>2025-03-20"
with patch.object(github, "_resolve_token", return_value="t"), patch.object(
github, "_fetch_json"
) as fetch:
result = github.search_github(topic, _FROM, _TO)
fetch.assert_not_called()
assert result["items"] == []
assert "error" not in result
def test_qualifier_only_logs_are_bounded_across_fanout():
topic = _long_qualifier_only_topic()
logs: list[str] = []
with patch.object(github, "_resolve_token", return_value="t"), patch.object(
github, "_fetch_json"
) as fetch, patch.object(github, "_log", side_effect=lambda m: logs.append(m)):
for _ in range(5):
github.search_github(topic, _FROM, _TO)
fetch.assert_not_called()
assert logs
for msg in logs:
assert len(msg) < _ERROR_BOUND
assert topic not in msg
def test_mixed_topic_strip_log_is_bounded():
subject = "open source ai " * 200
topic = subject + "stars:>1000 created:>2025-03-20"
logs: list[str] = []
with patch.object(github, "_resolve_token", return_value="t"), patch.object(
github, "_fetch_json", return_value={"items": []}
), patch.object(github, "_log", side_effect=lambda m: logs.append(m)):
github.search_github(topic, _FROM, _TO)
strip_logs = [m for m in logs if m.startswith("Stripped search qualifiers:")]
assert strip_logs
for msg in strip_logs:
assert len(msg) < _STRIP_LOG_BOUND
assert subject not in msg