CVE-2022-36804: Atlassian Bitbucket command injection, remote code execution, actively exploited
The short version: CVE-2022-36804 is a command-injection flaw in multiple Bitbucket Server and Data Center API endpoints. An attacker who can read a repository — and a public repository needs no account — can execute arbitrary code by sending a malicious HTTP request. It's rated 8.8, it's in CISA's KEV catalog, and it's easily exploitable. If you self-host Bitbucket, patch and verify. Steady hands, but move.
At a glance
| Fact | Detail |
|---|---|
| Our severity take | High — in practice (see below) |
| CVSS v3.1 (NVD) | 8.8 — High · CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H |
| EPSS | 99.08% · 99.93rd percentile (2026-07-17) |
| In CISA KEV? | Yes — remediation was due 2022-10-21 |
| Known exploited? | Yes — actively exploited; public exploit code available |
| Vulnerability type | CWE-78 OS command injection (with CWE-88 argument injection) → remote code execution |
| Requires authenticated session? | Read access to a repository — no account needed for a public repo |
| Affected | Bitbucket Server & Data Center: 7.0.0–7.6.16; 7.7.0–7.17.9; 7.18.0–7.21.3; 8.0.0–8.0.2; 8.1.0–8.1.2; 8.2.0–8.2.1; 8.3.0 |
| Fixed in | 7.6.17 · 7.17.10 · 7.21.4 · 8.0.3 · 8.1.3 · 8.2.2 · 8.3.1 (and later) |
What you need to know
Several Bitbucket API endpoints failed to properly neutralize input, allowing command injection — an attacker's request is turned into commands the server runs. The practical bar is low:
- Read access is enough, and if a repository is public, "read access" is available to anyone on the internet. That effectively turns this into an unauthenticated flaw on many instances.
- It's easily exploitable and public. Working exploit code and analysis were published shortly after disclosure, and it landed in CISA's KEV catalog within weeks.
- The target is your source-code platform. Code execution on Bitbucket reaches repositories, CI credentials, and the software supply chain behind them.
How serious we see it
High — in practice, slightly above the 8.8 base.
The base score assumes low privileges (a repo read), but on any instance that exposes a public repository, that privilege is free — the practical profile is unauthenticated RCE on a source-code server. Add confirmed in-the-wild exploitation and public tooling, and this is top-of-queue. The bounded, reassuring part: it affects specific Bitbucket builds, the fixes are out, and exposure is quick to confirm.
Recommendations
Straight from Atlassian's advisory and CISA:
- Patch now. Upgrade to 7.6.17 / 7.17.10 / 7.21.4 / 8.0.3 / 8.1.3 / 8.2.2 / 8.3.1 or later.
- Reduce exposure until patched. If you can't upgrade immediately, restrict access to the affected host to trusted networks, and reconsider which repositories are public.
- Hunt. Review request logs and the host for signs of command execution and implants.
- If compromised, respond. Rebuild, and rotate repository, CI/CD, and service credentials that the server could reach.
- Confirm your exposure first. Verify whether you run an affected self-hosted Bitbucket build.
How BreachRisk sees it
BreachRisk discovers internet-facing Bitbucket in your external footprint, fingerprints the product and version, and flags exposure tied to KEV-listed, actively exploited vulnerabilities like this one — surfaced at the top of your results because it's in KEV. We identify and flag the affected version; we don't exploit it.
The continuous, attacker's-eye view means your exposed source-code platform is already mapped when a flaw like this breaks, so you can move straight to patch-and-verify.