assafelovic / gpt-researcher
Code health
Hotspots average 2.0, the place fixes pay off most.
Explore this codebase
- Code health map7hotspots across 614 files
- Ownership205sole-owned files: one author, no second reader
- Architecture10modules, 23 entry points
- Decisions16recorded, of which 0 in force and 16 background
- Refactoring43open opportunities, one per file
- TestsNo coverage report yet. Run your tests with coverage on, then add the report:
AlsoDocsChatFilesKnowledge graphChange couplingCommitsStatsCosts
Do next · 3
3 things worth doing in the week to Jul 14, the last indexed commit.
Worth planning
Start breaking up create_chat_completion_with_tools (CCN 24, 109 lines)
gpt_researcher/utils/tools.py:158
create_chat_completion_with_tools nests deeply; easier to read and change once split; its file is imported by 1 file, changed 1 time in 90 days.
- gain
- up to +1.7 health on this file(inferred)
- finding
- Nested complexity
- size
- CCN 24, 109 lines, nests 8 deep
- severity
- high
Medium effort, worth a checkAddress the finding in <anonymous@1>
frontend/scripts.js:1
<anonymous@1> has a bumpy road finding; function changed in 18 commits; changed 1 time in 90 days.
- gain
- up to under +0.1 health on this file(inferred)
- finding
- Bumpy road
- size
- CCN 293, 1,803 lines, nests 4 deep
- severity
- high
Medium effort, worth a checkStart breaking up from_provider (CCN 41, 170 lines): first lift lines 206-213
gpt_researcher/llm_provider/generic/base.py:206
from_provider has many paths; easier to read and change once split; function changed in 29 commits; its file is imported by 10 files, changed 6 times in 90 days.
- gain
- up to +1.5 health on this file(inferred)
- finding
- Complex method
- size
- CCN 41, 170 lines
- severity
- high
Medium effort, worth a check
Improve what Repowise can see
Add a test coverage report
Without one, Repowise cannot tell tested code from untested code, so every test-related action says “unknown”.
- line coverage
- Unknown, no report
Small effort
Everything we found
5 areas
Every detector's open count by area. The actions above are the part worth doing.
- Code health731 open findingsCriticalUntested hotspot · gpt_researcher/skills/researcher.py35 findings · 9.6 deducted
- Refactoring43 open opportunitiesMedium<anonymous>+1.5 health, S effort · frontend/nextjs/public/workbox-f1770938.js
- Decisions0 drifting from the codeMediumgpt_researcher/skills/deep_research.pyHigh-churn file with no governing decision
- Ownership205 sole-owned filesMediumgpt_researcher/retrievers/utils.pyOne author, no second reader
- Dead code40Lowevals/hallucination_eval/run_eval.pyUnreachable symbol (90 lines)