Sourcery

The VS Code, Cursor, Windsurf or JetBrains extension reviews your uncommitted diff before anyone else sees it, and the same rules follow the change onto the pull request, so the bot's comment arrives while you can still fix it quietly. That is the daily trait that decides it: review moves left without a new tool to open. Public repositories are free; private ones need a paid seat from day one.

Pick it for editor-first review on a team that already argues about style. Pick CodeRabbit if the pull request is the only place you want comments, and Greptile if codebase context matters more than speed.

The risk is inconsistency. The docs say the status check can prevent merges, then note that GitLab does not receive status checks, so one configuration yields two policies: a gate on one host and a comment on the other, and a mixed shop discovers which is which when something merges that should not have. The failure mode is quiet by construction.

Treat the check as advisory everywhere until the GitLab side matches. What it does right: drafts, dependency-bot pull requests and packaging-only changes are skipped by default, so the bot does not spend its credibility on noise.

The documented pipeline is legible. 1. A summary of purpose and risk. 2. A reviewer guide that maps the change file by file and states how to verify it. 3. Inline comments with one-click apply. 4. A status check. Reviews run on open and again on each push, so the unit of review is the diff at that moment rather than the pull request as a whole. Nothing is executed; verification is a text the reviewer is asked to perform.

No benchmark is published for review precision or recall, so quality is asserted. The observation: it writes the test plan it cannot run.

Sourcery predates the agent wave; the first release is dated July 2019, and it has already pivoted once, from automated refactoring to review, which means the team has survived one model transition and a category change. The play now is bundling: secrets, SAST, IaC, dependency and license scanning ride on the same seat, so it competes with Snyk as much as with the review bots, and security budgets are stickier than developer-tool budgets.

Likely acquirer: a security vendor that wants a review surface, or GitLab. Position: hold; the bundle is the moat, and it is a modest one.

The demo is a bot summarizing a pull request. Procurement: Team is $24 a seat, so $1,440 a month for sixty, 20% less on annual, with 3x Pro's rate limits, which matters on a busy monorepo. GitHub Enterprise Server and self-hosted GitLab are supported, so our hosts are covered; self-hosting the reviewer itself is Enterprise only. SSO, SCIM, audit logs and retention are absent from the page.

CI fit is native, onboarding is an app install and a config file. Approved with conditions: those four items in writing before the seats, and a rate-limit test on the largest repo first.

I cannot read it or run it on my box; the extension proxies through their cloud even for uncommitted diffs, so the local review is local in name only. No MCP client, no MCP server, no local models, and a fork is out of the question because there is no source to fork.

The one handle: Team tier has bring your own LLM, and the list includes Azure OpenAI, so my tokens stay in a tenant I control and the vendor sees prompts but not my model bill. Grudging respect for the Azure option, which is more than most review bots allow, and no respect for the rest.