Borehole · Project · free tier
supabase/realtime
Free reading: not for reliance
90ccb30 · 3,207 commits · 745 files
Read 2026-10-04

supabase/realtime: technical due diligence, read from its repository

What Borehole read in supabase/realtime's code and history at 90ccb30: 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 Maturitystrong
Product & Engineering Maturitystrong
Security & Reliability Postureadequate
Observability & Operationsadequate
Fit with the Acquirernot assessable

What the repository says

  • CONTRIBUTING.md:38: states a policy on AI-generated contributions

Architecture & Codebase — including AI provenance

strong
01.6

forum/lib/forum/muster/scope.ex is 2,251 lines, 32× this codebase's median file

adequate

The median source file here is 70 lines. 4 files exceed 1,000 lines, the largest being 2,251. 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
forum/lib/forum/muster/scope.ex:1 — 2,251 lines (32× median)
priv/repo/tenant_db_dump_17.sql:1 — 1,517 lines (21× median)
priv/repo/tenant_db_dump_15.sql:1 — 1,516 lines (21× median)
lib/realtime_web/channels/realtime_channel.ex:1 — 1,272 lines (18× median)

Engineering Process & SDLC Maturity

strong

No finding.

Product & Engineering Maturity

strong

No finding.

Security & Reliability Posture

adequate
05.1

11 credential-shaped strings committed to the tree

adequate

Matches for JWT appear in tracked files outside test and fixture paths. They stay in the history even after they leave the tree, so what matters is whether each was rotated. The code cannot tell whether any of them is live, and that decides whether this matters: none sits in production configuration, and none is in a format its provider keeps for live keys. Is any of these live? Show that each was rotated after it was committed, and when. 14 other matchs set aside: expired tokens, licence tokens a vendor verifies with a public key, or placeholder text where a key would go. None grants access to anything.

Evidence
5 cited: JWT pattern matched
05.27

The build runs third-party code it does not pin

adequate

2 third-party CI actions are referenced by a tag or branch rather than a commit, so they run whatever that tag points to on the day; 1 step pipes a download straight into a shell. Tags have been moved to malicious commits in real supply-chain attacks, and a piped script is whatever the server returns that minute. Pinning to a commit is the fix, and it is small.

Evidence
.github/workflows/lint.yml:110 — action pinned by tag
.github/workflows/update-supabase-js.yml:55 — action pinned by tag
Dockerfile:16 — download piped into a shell
05.26

3 locked dependencies have published vulnerabilities

adequate

3 of 116 locked dependencies match 7 published advisories in OSV. A lockfile says which version is installed, not whether the flawed code is reachable from this product, so each needs a look. Most are fixed by an upgrade, and the advisory names the version. This lockfile pins development tools alongside what ships and does not say which is which, so some of these may never reach a user.

Evidence
mix.lock — cowlib 2.19.0: EEF-CVE-2026-43966 (moderate)
mix.lock — lazy_html 0.1.12: EEF-CVE-2026-92106 (unrated), fixed in 0.1.13
mix.lock — mint 1.10.1: EEF-CVE-2026-91043 (unrated), fixed in 1.10.2
05.65

4 advisories against this project in the last 24 months, 1 high or critical

adequate

The newest was published on 2026-09-24. A project that publishes advisories is telling its users about its fixes, and the count and pace say how much security work it carries. None of them affects the surveyed version. The record is incomplete: NVD: 51 records name realtime with nothing tying them to this repository (CVE-2014-7506, CVE-2015-7439, CVE-2019-19789, …).

Evidence
2026-09-24: GHSA-9vjf-j9f7-j42c (high): Client-supplied bytes are spliced unvalidated into other subscribers' JSON frames, letting any broadcaster forge Realtime protocol events (including postgres_ch
2026-08-17: GHSA-pfwq-rf2r-mjgc (low), fixed in 2.128.3: Unauthenticated log injection / terminal escape injection via OpenTelemetry baggage header
2026-06-25: GHSA-rcr8-2525-4r7p (moderate), fixed in >= 2.111.2: Incorrect Authorization in supabase/realtime
2026-06-16: GHSA-hg9h-6mhq-6c2v (low), fixed in >= 2.103.1: Expired JWTs with string exp are accepted by REST broadcast authorization

Observability & Operations

adequate
06.8

One request cannot be followed through the logs

adequate

Nothing in the service carries a request or correlation id, and there is no tracing. When a customer reports a failure, the log lines that belong to their request cannot be picked out from everyone else's.

Evidence
no request id, correlation id or tracing found
06.9

No error reporter is wired in

adequate

Nothing in the service initialises an error reporter (Sentry, Rollbar, Bugsnag, Honeybadger or an APM agent). An exception in production reaches a person only if someone is reading the logs, which usually means a customer reports it first.

Evidence
no error reporter initialised

Fit with the Acquirer

not assessable

Not rated: structurally not assessable from a repository.

Read more, or read your own

The full report on supabase/realtime: 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.