The most accurate, cheapest and fastest memory for coding agents: searchable session history from Claude Code, Codex, Cursor and 32 more agents, already on your disk. No LLM, one Go binary.
C-Rank Grade: S (Elite) - 75/100
External PR Merge Rate: 72%
Response Time: 8h
First Timer Success: 75%
vshulcz/deja-vu has a recorded first-timer success rate of 75.0%. Repositories ranked S typically provide actionable feedback during code reviews and actively nurture new community contributors.
Maintainers in vshulcz/deja-vu respond to incoming external pull requests in approximately 8.0 hours on average. Keeping PRs focused on single tasks and ensuring tests pass helps maintainers review faster.
The C-Rank™ system evaluates GitHub projects on a 0–100 scale using real data: PR merge rates, review turnaround time, active maintainer presence, and first-time contributor success. A score of 74.7 places vshulcz/deja-vu in the S tier.
The external contributor pull request merge rate for vshulcz/deja-vu is 72.0%, based on public PR activity from non-core contributors.
vshulcz/deja-vu currently has 3 active issue(s) tagged with beginner-friendly labels like "good first issue", "beginner", or "up-for-grabs".
Growing Go project in vshulcz/deja-vu welcoming community pull requests and bug fixes.
Average Response Latency
Tracks hours until a maintainer leaves a review, comment, or PR response.
External Acceptance Rate
Percentage of community pull requests successfully merged into main.
First PR Conversion
Rate at which developers submitting their first repository PR succeed.
Show contributors that your repository actively reviews and merges external pull requests.
[](https://getmerged.abhishekco.de/vshulcz/deja-vu?utm_source=github&utm_medium=badge)
Historical health & responsiveness evolution
External PR acceptance
First PR success rate
50% receive response in
90% receive response in
Maintainer review activity heatmap & prime window prediction
Calculating local time...
Calculating local time...
Detected: UTC
Maintainers are most responsive around 11:00 UTC (11:00 UTC in your timezone). Recommended submission window is Monday - Wednesday from 09:00 to 13:00 UTC (09:00 - 13:00 UTC local).
Maintainer Strategy: Maintainer Sweet Spot: PRs under 150 lines merge in ~5 hours (90% approval). Large PRs (>500 lines) take ~38 hours (32% approval). Keep your diff compact for rapid review.
The block of code a session changed leaves out edits the harness refused (#4953), but only for Claude Code and Codex. The generic reader used for every other harness (internal/sources/changes_generic.go) takes every edit call it finds, including one Gemini CLI marks "status":"error". Such a call changed nothing and shows as code the session wrote. Wanted: when the call object around the arguments carries a status that says it failed (error, cancelled), its edit is skipped. The walker already knows which call it is inside (walk passes the call name down to edit, changes_generic.go:159); it needs that call's status too. Test: add a failed replace beside a successful one to a Gemini-shaped session in internal/sources/changes_generic_test.go and check only the successful one comes back. Fails before the change, passes after.
Same as #4971, for Qwen Code. Qwen writes tool calls as functionCall parts with name and args (internal/sources/qwen.go:286), which is the shape the generic reader (#4955) looks for, but no Qwen session with an edit has been through it. Wanted: In a real Qwen Code, have it edit one file and create another. deja ctx <id> on that session should show both in ## what it changed. Add the trimmed shape to fixtures/registry/qwen/ and a test in internal/sources that reads the two changes back. TestChangesReadsAnotherHarnessesEditCalls in internal/sources/changes_generic_test.go:15 is the shape to copy. If the block comes back empty, the reason is almost always a field spelled differently from the lists at internal/sources/changes_generic.go:24-33; adding the spelling there is the fix. Keep the fixture small and make up its content.
Same as #4971, for Kimi Code. docs/registry/kimi.md says file edits are tool.call events with name and args: path with old_string/new_string on Edit, the whole file as content on Write. The generic reader (#4955) should read those, but no Kimi session with an edit has been through it. Wanted: In a real Kimi Code, have it edit one file and create another. deja ctx <id> on that session should show both in ## what it changed. Add the trimmed shape to fixtures/registry/kimi/ and a test in internal/sources that reads the two changes back. TestChangesReadsAnotherHarnessesEditCalls in internal/sources/changes_generic_test.go:15 is the shape to copy. If the block comes back empty, the reason is almost always a field spelled differently from the lists at internal/sources/changes_generic.go:24-33; adding the spelling there is the fix. Keep the fixture small and make up its content.
deja ctx <id> and MCP recall_context now show the code a session changed, ahead of the conversation (#4935). For Gemini CLI that comes from a generic reader (#4955), and it has only been checked against a hand-written session: a gemini turn whose toolCalls hold replace (file_path, old_string, new_string) and write_file (file_path, content). Wanted: In a real Gemini CLI, have it edit one file and create another. Run deja ctx <id> on that session and check the ## what it changed block shows both, paths relative to the project. Add the session's shape, trimmed, to fixtures/registry/gemini/ and a test in internal/sources that reads the two changes back. TestChangesReadsAnotherHarnessesEditCalls in internal/sources/changes_generic_test.go:15 is the shape to copy. If the block comes back empty, the reason is almost always a field spelled differently from the lists at internal/sources/changes_generic.go:24-33; adding the spelling there is the fix. Keep the fixture small and make up its cont
Twelve guide pages have a Chinese pair. docs/guide/delete-old-claude-sessions.html does not. Wanted: docs/zh/guide/delete-old-claude-sessions.html. Wanted: translated rather than machine-passed, in the voice the existing Chinese pages use (docs/zh/guide/find-a-session.html is the closest). What has to line up, and the tests that check it: The hreflang triple on both pages: zh-Hans, en and x-default, each side pointing at the other: TestChinesePagesArePairedWithTheirEnglishOnes. A row in docs/sitemap.xml and in docs/sitemap.txt: TestTextSitemapMatchesTheXMLOne. inLanguage: "zh-Hans" in the JSON-LD, and the sidebar links pointing inside docs/zh/guide/. A line in docs/llms.txt beside the other Chinese pages. Native or fluent Chinese only, please. Commands, flags and paths stay verbatim, as they do on the other Chinese pages.
Rate what actually matters after opening a pull request here.
Rate your first-hand PR experience (review speed, maintainer responsiveness, and onboarding ease) to help other contributors.
When evaluating whether to contribute to vshulcz/deja-vu, response velocity and maintainer engagement are crucial. GetMerged continuously tracks pull request trajectories, first-comment latency, and code review rounds to help developers avoid submitting pull requests to backlogged repositories.
Currently, maintainers of vshulcz/deja-vu acknowledge new external contributions in approximately 8 hours. Out of all submitted pull requests from non-core authors in the last 180-day window, 72.0% were successfully merged into the primary branch.
vshulcz/deja-vu has a recorded first-timer success rate of 75.0%. Repositories ranked Elite typically provide actionable feedback during code reviews and actively nurture new community contributors.
The initial maintainer response time averages ~8 hours. Keeping PRs scoped to single concerns and ensuring CI checks succeed will optimize review turnaround.
The C-Rank™ index scores repositories on a 0 to 100 scale using an objective formula: external PR merge rates, initial response speed, active maintainer count, and first-time contributor retention. A score of 74.7 places vshulcz/deja-vu in the Elite tier.
The external pull request merge rate is 72.0%. GetMerged isolates non-core community contributions so external developers get an accurate benchmark of PR acceptance probability.
Yes, vshulcz/deja-vu currently has 3 active issue(s) tagged with beginner-friendly labels. You can inspect these directly from the repository issues tab.
All metrics displayed for vshulcz/deja-vu are automatically retrieved via the public GitHub API and recalculated daily. Insider pull requests submitted by repository owners or organization members are excluded from merge rate calculations to preserve objective external contributor statistics.