Borehole · Project · free tier
infiniflow/ragflow
Free reading: not for reliance
a55dcce · 10,384 commits · 5,450 files
Read 2026-10-23

infiniflow/ragflow: technical due diligence, read from its repository

What Borehole read in infiniflow/ragflow's code and history at a55dcce: 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 provenanceadequate
Engineering Process & SDLC Maturitystrong
Product & Engineering Maturityadequate
Security & Reliability Postureadequate
Observability & Operationsstrong
Fit with the Acquirernot assessable

What the repository says

  • Releases are cut from another branch than the one surveyed

Architecture & Codebase — including AI provenance

adequate
01.6

internal/binding/cpp/re2/unicode_groups.cc is 6,513 lines, 42× this codebase's median file

adequate

The median source file here is 153 lines. 55 files exceed 1,530 lines, the largest being 6,513. 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
internal/binding/cpp/re2/unicode_groups.cc:1 — 6,513 lines (42× median)
internal/service/chat_pipeline.go:1 — 4,646 lines (30× median)
internal/cli/user_parser.go:1 — 4,538 lines (29× median)
web/src/locales/en.ts:1 — 4,222 lines (27× median)
web/src/locales/az.ts:1 — 4,146 lines (27× median)
01.25

1 base image floats: no version, or `latest`

adequate

1 of 7 `FROM` lines in product Dockerfiles name an image without a version, or with `latest`. Each build pulls whatever the registry serves that day, so the image that ships can change without a commit, and last month's build cannot be reproduced. Pin a version, or a digest.

Evidence
Dockerfile:11 — base image without a version

Engineering Process & SDLC Maturity

strong

No finding.

Product & Engineering Maturity

adequate
04.3

126 direct dependencies in a single manifest

adequate

web/package.json 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
web/package.json — 126 declared
go.mod — 91 declared
pyproject.toml — 10 declared
internal/channels/whatsapp-gateway/package.json — 2 declared
04.20

Fixes are 62% of recent work, up from 41%

adequate

In the last two quarters 62% of commits were fixes, against 41% 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
67%, 57%, 47%, 43%, 42%, 39%, 40%, 32%
04.62

Releases are cut from another branch than the one surveyed

adequate

The newest release, v0.27.3, is not on the branch this survey read; the newest release on it is v0.27.2. The shipped product may differ from what this report describes. Survey the release branch too.

Evidence
tag v0.27.3

Security & Reliability Posture

adequate
05.2

None of the standard security guard rails are configured

adequate

There is no disclosure policy, no automated dependency updates, no named code owners and no pre-commit configuration. None of these is load-bearing on its own. Their collective absence says that security posture here is whatever the current team happens to remember to do, which is not a posture that survives the team changing.

Evidence
checked 5 kinds of guard rail across 5,450 tracked files; 0 present
05.6

2 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
deps/Dockerfile — runs as root
05.25

4 environment files are committed

adequate

docker/.env, docker/.env.single-bucket-example, web/.env and more are tracked in git. Environment files are where credentials usually live, and committing one usually means nobody decided how secrets reach the running application. Whatever is in it is in the history for good. Check 05.1 says whether it holds anything credential-shaped.

Evidence
docker/.env — committed environment file
docker/.env.single-bucket-example — committed environment file
web/.env — committed environment file
05.27

The build runs third-party code it does not pin

adequate

6 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. 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/release-python.yml:74 — action pinned by tag
.github/workflows/release.yml:234 — action pinned by tag
05.13

10 outbound calls can wait forever

adequate

10 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
internal/rag/decisionhead/remote.go:63 — outbound call without a timeout
internal/syncer/connector/gmail.go:71 — outbound call without a timeout
internal/syncer/connector/gmail.go:224 — outbound call without a timeout
05.20

37 SQL statements built from strings

adequate

37 places in product code build SQL by formatting or joining strings rather than passing parameters. In each, what the code shows going in is a table or column name, a constant, or something the line does not show, not a request's value. Ask whether any value reaches these strings, rather than a parameter.

Evidence
internal/dao/migration.go:904 — SQL built from a string
internal/dao/migration.go:920 — SQL built from a string
internal/dao/migration.go:1038 — SQL built from a string
05.21

2 shell commands built from strings

adequate

2 places run a command through a shell (`shell=True`, `os.system`, `exec` with a template) rather than passing arguments. Where any part comes from outside, it is command injection. Ask where each one's input comes from.

Evidence
internal/harness/core/prebuilt/coding/coding.go:228 — command run through a shell
internal/harness/core/prebuilt/coding/coding.go:251 — command run through a shell
05.26

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

adequate

11 of 486 locked dependencies match 13 published advisories in OSV, 5 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
sdk/python/uv.lock — certifi 2025.10.5: GHSA-6q48-6c7w-9gw2 (critical), fixed in 2026.10.19
uv.lock — certifi 2026.1.4: GHSA-6q48-6c7w-9gw2 (critical), fixed in 2026.10.19
go.mod — github.com/gomarkdown/markdown v0.0.0-20260614204949-e08cff860f76: GHSA-cv23-7vc5-jfh7 (high); pinned to a commit or a branch, so whether it has the fix needs the dependency's own source
05.45

1 workload setting give a container the host

adequate

1 places in the deployment configuration run a container privileged or in the host's network, process or IPC namespace. A compromise of that container is a compromise of the machine, and most acquirers' security reviews block it. Node agents (network, storage, monitoring) are excluded.

Evidence
docker/docker-compose-base.yml:222 — privileged or host namespace
05.48

7 places insert HTML from a variable without sanitising it

adequate

7 places set HTML from a value that is neither a literal nor passed through a sanitiser. Where the value can come from a user, that is stored cross-site scripting, and it is the first thing a buyer's security review looks for in a frontend.

Evidence
web/public/iconfont.js:190 — HTML from a variable
web/src/components/generate-log-button.tsx:71 — HTML from a variable
web/src/components/originui/select-with-search.tsx:338 — HTML from a variable
05.9

2 vendored components with no licence

adequate

Third-party code is committed into this repository without any statement of the terms it arrived under. Unknown terms are harder to deal with than inconvenient ones: an awkward licence can be complied with, an absent one cannot. Each of these needs its origin established before the code changes hands.

Evidence
deps/Dockerfile — 1 tracked files, no licence file
deps/download_deps.py — 1 tracked files, no licence file
05.65

18 advisories against this project in the last 24 months, 12 high or critical

adequate

The newest was published on 2026-10-05. A project that publishes advisories is telling its users about its fixes, and the count and pace say how much security work it carries. Whether 12 of them are fixed in the code you ship could not be told from this history: its version is only known to be at least 0.27.2. Answer lives in: the release you run, set against each fixed version. The record is incomplete: NVD: 5 records name ragflow with nothing tying them to this repository (CVE-2024-12871, CVE-2024-12880, CVE-2026-51893, …).

Evidence
2026-10-05: GHSA-c6q8-23rp-32cv (high): OpenAI-compatible agent completion bypasses private canvas authorization
2026-10-01: CVE-2026-51892 (moderate): infiniflow ragflow 0.24.0 is vulnerable to Incorrect Access Control via /v1/document/get/<doc_id>.
2026-10-01: CVE-2026-51895 (moderate): Ragflow 0.24.0 and prior contains improper access control in update_metadata_setting (api/apps/kb_app.py). Depending on the exposed entry, an attacker can perfo
2026-10-01: CVE-2026-51896 (moderate): infiniflow ragflow 0.25.3 contains improper access control in resume (api/apps/connector_app.py). Depending on the exposed entry, an attacker can perform unauth
2026-09-17: CVE-2026-93013 (moderate): RAGFlow through 0.27.2 contains a path traversal vulnerability in the dev_insert_chunks_from_file and dev_insert_metadata_from_file endpoints that allows authen

Observability & Operations

strong
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 18 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/docker-compose-base.yml:163 — image without a version

Fit with the Acquirer

not assessable

Not rated: structurally not assessable from a repository.

Read more, or read your own

The full report on infiniflow/ragflow: 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.