Hankanman / Meetily-Local
Code health
This codebase scores 6.8 out of 10 for code health, which we rate fair. It also scores maintainability 7.0 and static performance 9.6 out of 10. The three are scored separately and never blended into one number. Risk is concentrated, as it usually is: 51 of 291 files are git hotspots, and they average 5.1 — which is where the fixes pay off most.
Do next
3 suggested
3 things worth doing in the week to Oct 19, the last indexed commit, 3 of them now.
Now
- High
Simplify
ask_acrossinparley-core/src/chat/across/mod.rswhile the change is fresh1 commit in the last 7 days added or worsened 1 critical or high finding here, still open. Fixing now costs less than after the next change.
- serious findings
- 1
- critical
- 0
- latest commit
- Stop the summary helper dying on a dead stderr pipe
Small effort - High
Simplify
watch_onceinparley-core/src/audio/meeting_detect/watcher.rswhile the change is fresh1 commit in the last 7 days added or worsened 1 critical or high finding here, still open. Fixing now costs less than after the next change.
- serious findings
- 1
- critical
- 0
- latest commit
- Stop the summary helper dying on a dead stderr pipe
Small effort - High
Simplify
retrieveinparley-core/src/chat/across/retrieval.rswhile the change is fresh1 commit in the last 7 days added or worsened 1 critical or high finding here, still open. Fixing now costs less than after the next change.
- serious findings
- 1
- critical
- 0
- latest commit
- Stop the summary helper dying on a dead stderr pipe
Small effort
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
- Lines of code
- 69.5K
Recent activity
All commitsCommits
- 370b8d03Stop live action items flagging every other sentence2h ago
- 45b979f1Make a voice match to someone not invited much less likely6d ago
- 771c4b82Let a wrong speaker name be unassigned, and relearn the voice6d ago
- 7bb424e9Stop the summary helper dying on a dead stderr pipe6d ago
- d2e8a3e9Merge speakers less aggressively to fit the attendee count1w ago
- 360f7d7cLet one transcript line be reassigned without its whole cluster1w ago
Decisions
- Rename meetings from summaries only when title is user-generated placeholder/emptyproposed · 2h ago
- Fix UTF-8 truncation panics in API response/token loggingproposed · 2h ago
- Merge duplicate voice profiles into the self profileproposed · 2h ago
- Add ignored end-to-end test for scratch-data retranscriptionproposed · 2h ago
- Score duplicate-named profiles as self to enforce system-channel ruleproposed · 2h ago
- Persist batch channel tags and refine mic cluster mergingproposed · 2h ago
Everything we found
1,194 open
Every detector's open count by area, each with its worst example. An inventory, not a to-do list: the actions above are the part worth doing.
- worst item is critical severity. Code health869God class · parley-gpui/src/views/meeting/mod.rs29 findings · 7.9 deducted
- worst item is low severity. Refactoring97MeetingView+0.9 health, XL effort · parley-gpui/src/views/meeting/mod.rs
- worst item is medium severity. Doc drift5docs/transcription-backends.mdLink points at #implementation-plan, which docs/transcription-backends.md no longer declares.
- worst item is medium severity. Decisions80parley-gpui/src/views/meeting/mod.rs29 awaiting review · 51 hotspots ungoverned
- worst item is medium severity. Ownership143parley-gpui/src/views/meeting/mod.rs100% single-owner concentration, 24 commits in 90d
Composition
Open the graph- rust98%
- shell2%
- python<1%
Explore this codebase
- Docs503 pages across 11 modules
- ChatAsk this codebase a question and get an answer with its sources.
- Files291 files with per-file docs, health and history
- ArchitectureDependency graph, layers, and 12 entry points
- Code healthPer-file scores, 869 open findings, coverage and refactoring targets
- Knowledge graphEntities, communities, and the paths between them
- Change couplingFiles that keep changing together, mined from commit history
- CommitsChange-risk ranked history with agent provenance
- ContributorsBus factor, per-file maintainers, and the human/agent split
- StatsSize class, origin, lifetime churn, rhythm and records
- CostsWhat indexing this snapshot cost, by model and by run.