Filtered view · Verified 2026-08-11
4 of the 27 tools tracked here publish their source, which means you can read how the review prompt is built, audit what leaves your network, and keep running it if the vendor's terms change.
Two things the badge does not tell you: whether it is self-hostable, and what the licence obliges you to do. Both are in the table.
| # | Tool | Coverage | Self-hosting | Licence | Free tier | From |
|---|---|---|---|---|---|---|
| 1 | Kodus AI PR Review | 6.5/9 Coverage score 6.5 out of 9 | Self-hostable | Open source | Limited free tier | $10/dev/mo |
| 2 | Semgrep Security | 4.5/9 Coverage score 4.5 out of 9 | Self-hostable | Open source | Limited free tier | $30/contributor/mo |
| 3 | SonarQube Code Quality | 4/9 Coverage score 4 out of 9 | Self-hostable | Open source | Limited free tier | $34/mo |
| 4 | PR-Agent AI PR Review | 2.5/9 Coverage score 2.5 out of 9 | Self-hostable | Open source | Free tier | Free (MIT) |
AI PR Review · AGPL-3.0
Open-source AI code review built to run where the organization controls: self-hosted or cloud, models under your keys, org-wide Kody Rules.
Security · LGPL-2.1
Open-source static analysis engine with a commercial AppSec Platform adding cross-file analysis and an AI assistant for triage.
Code Quality · LGPL-3.0
Deterministic static analysis platform (Server and Cloud) with PR decoration and LLM-generated AI CodeFix suggestions.
AI PR Review · MIT
Community-maintained, MIT-licensed PR reviewer with /review, /describe, and /improve commands, self-hosted with your own model keys.
4 of the 27 tools in this directory publish their source: Kodus (AGPL-3.0), Semgrep (LGPL-2.1), SonarQube (LGPL-3.0), PR-Agent (MIT).
Not in practice. The review engine may be free to run, but the model calls are not — with an open-source tool you pay the LLM provider directly instead of paying a vendor margin. Some projects are also dual-licensed, with enterprise features under a commercial licence.
AGPL requires you to publish changes if you offer the modified software as a network service; MIT and Apache do not. For a tool you run internally the practical difference is usually small, but it matters to your legal review and it matters if you plan to build a product on top.
On documented coverage of the 9 standards they are competitive but not automatically ahead: the open-source tools here score between 2.5 and 6.5 out of 9, against 1–7.5 for the proprietary ones. Read each matrix rather than assuming the licence predicts the quality of findings.
Need it inside your own network?
Open source and self-hostable overlap, but they are not the same list.