Dataset Viewer
Auto-converted to Parquet Duplicate
run_id
string
created_at
string
repo
string
issue_title
string
issue_body
string
status
string
iterations
int64
findings_count
int64
findings
list
repro_test_path
string
repro_test
string
fix_rationale
string
patch_unified_diff
string
commit_sha
string
initial-benchmark-run
2026-11-19T17:53:03.604198+00:00
calc.py
Off-by-one denominator in calc.average(xs)
Divides by len(xs)+1 instead of len(xs)
committed
1
1
[ { "severity": "High", "file_path": "calc.py", "title": "Off-by-one denominator in average()", "description": "Divides by len(xs)+1", "suggested_fix": "Use len(xs)" } ]
5e9f468
live-4b3698c4
2026-11-19T19:03:58.273507+00:00
/tmp/bugfix-clone-v9gbv1nt
backend/agent.py – clean_json_response returns None instead of cleaned JSON string
In the function `backend.agent.clean_json_response`, when called with a string that contains a JSON code block (e.g., "```json\n{\"key\": \"value\"}\n```"), the function extracts the inner JSON text but never returns it. The current behavior is that the function returns `None`, causing downstream code that expects a JS...
escalated
0
7
[ { "severity": "Critical", "file_path": "backend/agent.py", "title": "clean_json_response does not return the cleaned JSON string", "description": "The function extracts JSON from markdown code fences but never returns the cleaned text, resulting in `None` being propagated to callers. This leads to d...
test_repro_bug.py
import sys import types import importlib # Prepare a lightweight dummy module for backend.rag_engine to avoid heavy imports dummy_rag_engine = types.ModuleType("backend.rag_engine") # The agent only needs the `search_similar_chunks` symbol at import time dummy_rag_engine.search_similar_chunks = lambda *args, **kwargs:...
null

No dataset card yet

Downloads last month
-