Borehole · Project · free tier
owasp-modsecurity/modsecurity
Free reading: not for reliance
1ebdc39 · 4,046 commits · 856 files
Read 2026-11-03

owasp-modsecurity/modsecurity: technical due diligence, read from its repository

What Borehole read in owasp-modsecurity/modsecurity's code and history at 1ebdc39: its licence, security advisories, releases, review practice, tests, dependencies and operations. Every finding cites the file and line it rests on. The full report also covers the team and asks the questions the code cannot answer.

DimensionRating
Architecture & Codebase — including AI provenancestrong
Engineering Process & SDLC Maturityadequate
Product & Engineering Maturityadequate
Security & Reliability Postureadequate
Observability & Operationsnot assessable
Fit with the Acquirernot assessable

What the repository says

  • LICENSE: Relicensed: GPL-2.0 → Apache-2.0 (2011-03-30)
  • .gitmodules:1: 4 submodule(s) this survey does not read: test/test-cases/secrules-language-tests, others/libinjection, bindings/python

Architecture & Codebase — including AI provenance

strong
01.6

src/parser/seclang-scanner.cc is 10,192 lines, 216× this codebase's median file

adequate

The median source file here is 47 lines. 8 files exceed 1,000 lines, the largest being 10,192. Size is not a defect on its own — some problems genuinely live in one place — but these are where merge conflicts, review fatigue and single-owner knowledge concentrate, and they are the first thing that slows an inheriting engineer down.

Evidence
src/parser/seclang-scanner.cc:1 — 10,192 lines (216× median)
src/parser/seclang-parser.hh:1 — 8,950 lines (190× median)
src/parser/seclang-parser.cc:1 — 7,503 lines (159× median)
src/parser/seclang-parser.yy:1 — 3,184 lines (67× median)
src/transaction.cc:1 — 2,390 lines (50× median)

Engineering Process & SDLC Maturity

adequate
02.3

Tests cover a thin slice of the codebase

adequate

42 test files against 531 source files (7.9%). This does not measure whether the tests are good, only whether enough of them exist for the codebase to be changed safely by someone who did not write it — which is exactly the situation after an acquisition.

Evidence
42 test files / 531 source files
02.6

Almost no process scaffolding exists in the repository

adequate

Of five ordinary markers — CI workflows, CODEOWNERS, a pull-request template, pre-commit hooks, contribution guidance — 1 is present: CI workflows. Branch protection itself is a GitHub setting rather than a file, so this is a reading of proxies and not of the rules themselves. What it does establish is that the process is not written down anywhere a new engineer would find it.

Evidence
1 of 5 markers present across 856 tracked files

Product & Engineering Maturity

adequate
04.5

Effort is shifting from building to rewriting

adequate

Deleted lines per line added rose from 0.45 to 0.93 between the first and second halves of this history. Rework is normal and often healthy, but a rising ratio means an increasing share of the team's time is spent replacing code that already exists rather than building what does not. Whether that is deliberate repayment or churn is a question for the team.

Evidence
first half 0.45, second half 0.93 (split 2016-12-19)
04.20

Fixes are 52% of recent work, up from 23%

adequate

In the last two quarters 52% of commits were fixes, against 23% in the quarters before. Rising fix work usually means quality debt is coming due, and it takes time that new work would have had.

Evidence
39%, 65%, 22%, 16%, 25%, 26%, 13%, 39%

Security & Reliability Posture

adequate
05.6

1 Dockerfile runs the application as root

adequate

No USER instruction switches away from root, so the process runs with full privileges inside the container. It is a one-line fix and an ordinary expectation, which is why its absence is worth asking about: it usually says more about whether anyone has reviewed the deployment than about this container specifically.

Evidence
.oss-scanner/Dockerfile — runs as root
05.60

Relicensed: GPL-2.0 → Apache-2.0 (2011-03-30)

adequate

The root licence changed 1 time(s). Code taken under earlier terms stays under them; check which terms any fork or customer relied on.

Evidence
LICENSE — GPL-2.0 from 2007-02-06
LICENSE — Apache-2.0 from 2011-03-30
05.65

17 advisories against this project in the last 24 months, 9 high or critical

adequate

The newest was published on 2026-09-30. A project that publishes advisories is telling its users about its fixes, and the count and pace say how much security work it carries. The record is incomplete: NVD: 33 records name modsecurity with nothing tying them to this repository (CVE-2008-5676, CVE-2009-1902, CVE-2009-1903, …).

Evidence
2026-09-30: GHSA-5pww-8rfg-9crf (high): RFC 2231 `filename*` Parameter Bypasses Multipart Filename Rules on RFC-Compliant Backends (Go, Python, Node, Java)
2026-09-30: GHSA-4j47-8qcr-jf59 (moderate), fixed in 2.9.15: `t:base64DecodeExt` silently discards the entire decoded value on a malformed base64 padding group, bypassing rule matching
2026-09-30: GHSA-qrch-pjfr-9g47 (moderate), fixed in 2.9.15: removeComments Transformation Fails to Strip Adjacent Comments, Enabling WAF Rule Bypass
2026-09-30: GHSA-5m93-4h75-3p2w (moderate), fixed in 3.0.17: PCRE2 `@rxGlobal` match-limit errors are converted to no-match
2026-09-30: GHSA-jx3r-phvx-2jmj (high), fixed in 3.0.17: ModSecurity v3 XML Request Body Processor Uninitialized Pointer Dereference Vulnerability

Observability & Operations

not assessable

Not rated: only 4 of 16 checks could reach a conclusion here; the rest do not apply to this repository or could not read it.

Fit with the Acquirer

not assessable

Not rated: structurally not assessable from a repository.

Read more, or read your own

The full report on owasp-modsecurity/modsecurity: every dimension, the questions for the room, and the audit log.

Survey a repository: free on public repositories. Private code is read on your own machine by a collector whose source you can read first.