Borehole · Project · free tier
argoproj/argo-cd
Free reading: not for reliance
6cd2245 · 11,572 commits · 6,053 files
Read 2026-10-04

argoproj/argo-cd: technical due diligence, read from its repository

What Borehole read in argoproj/argo-cd's code and history at 6cd2245: 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 Maturityadequate
Security & Reliability Postureadequate
Observability & Operationsadequate
Fit with the Acquirernot assessable

What the repository says

  • Releases are cut from another branch than the one surveyed
  • AGENTS.md:13: names CNCF
  • README.md:10: names OpenSSF
  • SECURITY.md:72: states a policy on AI-generated contributions

Architecture & Codebase — including AI provenance

strong
01.6

gitops-engine/pkg/utils/kube/scheme/parser.go is 17,708 lines, 276× this codebase's median file

adequate

The median source file here is 64 lines. 33 files exceed 1,000 lines, the largest being 17,708. 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
gitops-engine/pkg/utils/kube/scheme/parser.go:1 — 17,708 lines (276× median)
pkg/apis/application/v1alpha1/openapi_generated.go:1 — 8,625 lines (134× median)
pkg/apis/application/v1alpha1/types.go:1 — 4,281 lines (66× median)
reposerver/repository/repository.go:1 — 3,772 lines (58× median)
pkg/apiclient/application/application.pb.gw.go:1 — 3,744 lines (58× median)

Engineering Process & SDLC Maturity

strong

No finding.

Product & Engineering Maturity

adequate
04.3

126 direct dependencies in a single manifest

adequate

go.mod declares 126 direct dependencies. Each one is a maintenance obligation, a supply-chain surface and a potential licence question that transfers with the asset. The count says nothing about whether any given dependency was a good choice; it says how many such choices a buyer inherits.

Evidence
go.mod — 126 declared
ui/package.json — 39 declared
gitops-engine/go.mod — 25 declared
04.38

8 files still carry code deprecated more than two years ago

adequate

8 files were marked deprecated over two years ago and still are, the oldest since Oct 2018. A deprecation is a promise to remove. Ask what still depends on these, and whether removing them is scheduled.

Evidence
pkg/client/informers/externalversions/factory.go — deprecated since Oct 2018
pkg/apiclient/repository/repository.pb.go — deprecated since Oct 2019
server/repository/repository.go — deprecated since Oct 2019
pkg/apis/application/v1alpha1/types.go — deprecated since Dec 2019
pkg/apis/application/v1alpha1/openapi_generated.go — deprecated since Dec 2019
04.62

Releases are cut from another branch than the one surveyed

adequate

The newest release, v3.5.3, is not on the branch this survey read; the newest release on it is v0.8.0. The shipped product may differ from what this report describes. Survey the release branch too.

Evidence
tag v3.5.3

Security & Reliability Posture

adequate
05.1

9 credential-shaped strings in the tree sit in tests, examples, documentation or local development settings

adequate

Strings shaped like credentials appear only where fakes are usual: test fixtures, examples, documentation, placeholders and local stacks. They are almost always fakes. Ask someone to confirm that none of them works anywhere, because a real key in a development file is still a real key. 10 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
Slack token, where fakes are usual
4 cited: JWT, where fakes are usual
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/image.yaml:145 — action pinned by tag
.github/workflows/release.yaml:210 — action pinned by tag
.github/workflows/ci-build.yaml:621 — download piped into a shell
05.13

3 outbound calls can wait forever

adequate

3 HTTP calls use a client with no timeout (Python `requests` without `timeout=`, or Go's default client). When a dependency slows down, each one holds a worker until the connection gives up, and enough of them take the service down with the dependency.

Evidence
applicationset/generators/scm_provider.go:325 — outbound call without a timeout
applicationset/utils/utils.go:632 — outbound call without a timeout
pkg/apiclient/apiclient.go:302 — outbound call without a timeout
05.34

25 of 25 Kubernetes workloads declare no resource limits

adequate

25 workload manifests set no CPU or memory limits. One pod that leaks memory or spins can starve everything else on its node, and the scheduler cannot plan capacity for what it cannot see.

05.26

18 locked dependencies have published vulnerabilities, 4 high or critical in what ships

adequate

18 of 1,568 locked dependencies match 41 published advisories in OSV, 19 of them rated high or critical. 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.

Evidence
go.mod — github.com/go-git/go-git/v5 v5.19.1: GHSA-hc8v-wwc9-vgxm (high), fixed in 5.19.2
ui/pnpm-lock.yaml — form-data 4.0.5: GHSA-hmw2-7cc7-3qxx (high), fixed in 4.0.6
ui/pnpm-lock.yaml — linkify-it 3.0.3: GHSA-22p9-wv53-3rq4 (high), fixed in 5.0.1
05.65

13 advisories against this project in the last 24 months, 10 high or critical

adequate

The newest was published on 2026-08-29. A project that publishes advisories is telling its users about its fixes, and the count and pace say how much security work it carries. GHSA-786q-9hcg-v9ff (critical), GHSA-2hj5-g64g-fp6p (critical), GHSA-47g2-qmh2-749v (moderate) are fixed in the surveyed commit; which release carries them cannot be told from this history, which does not place the release users run. Answer lives in: the version you run. The record is incomplete: NVD: 1 record names argo-cd with nothing tying it to this repository (CVE-2026-14251).

Evidence
2026-08-29: CVE-2026-82456 (critical): argocd-mcp 0.8.0 binds its HTTP transport to every network interface and accepts MCP sessions without requiring caller credentials when ARGOCD_API_TOKEN is conf
2026-07-14: CVE-2026-15416 (high): A flaw was identified in Argo CD, the GitOps engine used by Red Hat OpenShift GitOps, that could allow an unauthenticated attacker with network access to the Ar
2026-07-13: CVE-2026-62185 (high): Argo CD Helm Chart before 10.0.0 fails to install network policies by default, allowing any pod on a cluster to access repo-server and other Argo APIs. Attacker
2026-05-13: GHSA-h98r-wv3h-fr38 (high), fixed in 3.2.12: Argo CD: Stored XSS in application link annotations enables developer-to-admin privilege escalation
2026-05-13: GHSA-rg3g-4rw9-gqrp (moderate), fixed in 3.2.12: Argo CD: Kubernetes Secret Extraction via ArgoCD ServerSideDiff via sensitive annotations

Observability & Operations

adequate
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
06.18

3 of 8 deployed images carry no version, or `latest`

adequate

3 images in the deployment configuration are named without a version or as `latest`. A redeploy can change the software without a commit, a rollback has nothing to roll back to, and nobody can say from the repository what is running.

Evidence
gitops-engine/agent/manifests/install.yaml:56 — image without a version
manifests/namespace-install.yaml:2901 — image without a version
manifests/dev-tilt/ui-deployment.yaml:16 — image without a version
06.14

Readiness and liveness probes are identical in 2 of 15 workloads

adequate

Both probes call the same endpoint. When a dependency slows down, the liveness probe fails with the readiness probe, so the platform restarts pods that only needed taking out of rotation, and a partial outage becomes a full one. Liveness should ask whether the process is alive; readiness whether it can serve.

Evidence
manifests/ha/namespace-install-with-hydrator.yaml — same probe endpoint twice
manifests/ha/namespace-install.yaml — same probe endpoint twice
06.33

6 of 25 deployments run a single replica

adequate

6 deployments set `replicas: 1`, and nothing (an autoscaler, a disruption budget) adds more. Every deploy and every node drain is downtime, and the availability the buyer inherits has a ceiling.

Fit with the Acquirer

not assessable

Not rated: structurally not assessable from a repository.

Read more, or read your own

The full report on argoproj/argo-cd: 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.