Borehole · Project · free tier
chroma-core/chroma
Free reading: not for reliance
52ce844 · 4,679 commits · 2,188 files
Read 2026-10-22

chroma-core/chroma: technical due diligence, read from its repository

What Borehole read in chroma-core/chroma's code and history at 52ce844: 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

Architecture & Codebase — including AI provenance

strong
01.6

rust/rust-sysdb/src/spanner.rs is 9,752 lines, 81× this codebase's median file

adequate

The median source file here is 120 lines. 68 files exceed 1,200 lines, the largest being 9,752. 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
rust/rust-sysdb/src/spanner.rs:1 — 9,752 lines (81× median)
rust/log-service/src/lib.rs:1 — 8,638 lines (71× median)
rust/types/src/collection_schema.rs:1 — 7,820 lines (65× median)
rust/frontend/src/impls/service_based_frontend.rs:1 — 5,995 lines (49× median)
rust/segment/src/blockfile_metadata.rs:1 — 5,684 lines (47× median)

Engineering Process & SDLC Maturity

strong

No finding.

Product & Engineering Maturity

strong

No finding.

Security & Reliability Posture

adequate
05.6

3 Dockerfiles run 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
Dockerfile — runs as root
go/Dockerfile — runs as root
rust/Dockerfile — runs as root
05.27

The build runs third-party code it does not pin

adequate

15 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; 2 steps pipe 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/_build_js_bindings.yml:23 — action pinned by tag
.github/workflows/_build_release_container.yml:103 — action pinned by tag
Dockerfile:14 — download piped into a shell
05.13

2 outbound calls can wait forever

adequate

2 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
chromadb/cli/cli.py:25 — outbound call without a timeout
sample_apps/generative_benchmarking/functions/embed.py:78 — outbound call without a timeout
05.26

89 locked dependencies have published vulnerabilities, 79 high or critical in what ships

adequate

89 of 2,786 locked dependencies match 196 published advisories in OSV, 124 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. 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
clients/new-js/pnpm-lock.yaml — convict 6.2.4: GHSA-44fc-8fm5-q62h (critical), fixed in 6.2.5
sample_apps/movies/package-lock.json — next 16.0.10: GHSA-2464-fhr4-w9xw (critical), fixed in 16.4.1
clients/new-js/pnpm-lock.yaml — tar 7.4.3: GHSA-23hp-3jrh-7fpw (critical), fixed in 7.5.19
05.65

7 advisories against this project in the last 24 months, 7 high or critical

adequate

The newest was published on 2026-09-16. 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: 4 records name chroma with nothing tying them to this repository (CVE-2020-16602, CVE-2024-29404, CVE-2026-28020, …).

Evidence
2026-09-16: CVE-2026-92782 (high): Chroma through 1.5.9 fails to validate tenant and database segments when resolving collections, allowing authenticated attackers to access collections from othe
2026-09-04: CVE-2026-85664 (high): Chroma 1.5.9 fails to validate maximum bounds on HNSW index parameters max_neighbors, ef_construction, and ef_search in collection-create requests. Unauthentica
2026-06-12: GHSA-2wm9-hf6c-p5cr (high): ChromaDB allows any authenticated users to arbitrarily read, write, update, or delete data in any tenant's collection
2026-06-12: GHSA-36p7-vc44-83pf (critical): ChromaDB has a code injection vulnerability
2026-06-12: GHSA-xph7-9rjv-w5fr (high): ChromaDB's SimpleRBACAuthorizationProvider doesn't check which tenant, database, or collection a permission applies to

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

1 of 1 deployed images carry no version, or `latest`

adequate

1 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
docker-compose.server.example.yml:8 — image without a version
06.14

Readiness and liveness probes are identical in 2 of 2 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
k8s/distributed-chroma/templates/foundation-service.yaml — same probe endpoint twice
k8s/distributed-chroma/templates/mdac-service.yaml — same probe endpoint twice

Fit with the Acquirer

not assessable

Not rated: structurally not assessable from a repository.

Read more, or read your own

The full report on chroma-core/chroma: 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.