Borehole · Project · free tier
iflytek/astron-agent
Free reading: not for reliance
468a676 · 3,257 commits · 4,913 files
Read 2026-10-23

iflytek/astron-agent: technical due diligence, read from its repository

What Borehole read in iflytek/astron-agent's code and history at 468a676: 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

  • README.md:182: names CNCF

Architecture & Codebase — including AI provenance

strong
01.3

Commit messages are templated rather than written

adequate

36% of commit subjects are exact repeats of another subject, and subject length varies by only 130.8 characters. A log written by someone explaining their reasoning to a future reader does not look like this. The log is therefore weak evidence of intent, whatever the code is like.

Evidence
79× 'cicd'
24× 'feat: Added skill system'
16× 'docs:Optimize deployment documentation'
01.6

console/frontend/src/utils/avatar-sdk-web_3.1.2.1002/xrtc-player-BJTnVhG9.js is 25,440 lines, 403× this codebase's median file

adequate

The median source file here is 63 lines. 49 files exceed 1,000 lines, the largest being 25,440. 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.

Engineering Process & SDLC Maturity

strong

No finding.

Product & Engineering Maturity

adequate
04.3

102 direct dependencies in a single manifest

adequate

core/workflow/pyproject.toml declares 102 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.

04.17

2 pinned runtimes are past end of life

adequate

The code is pinned to Go 1.23 (end of life 2025-08-12); Node.js 18 (end of life 2025-04-30). None of these receives security fixes any more; the first stopped on 2025-04-30. Moving off an unsupported runtime is rarely a one-line change, so a buyer should ask whether the upgrade is planned and what it touches.

Evidence
core/tenant/Dockerfile — Go 1.23, end of life 2025-08-12
console/frontend/Dockerfile — Node.js 18, end of life 2025-04-30

Security & Reliability Posture

adequate
05.6

10 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
console/backend/hub/Dockerfile — runs as root
console/frontend/Dockerfile — runs as root
core/agent/Dockerfile — runs as root
core/knowledge/Dockerfile — runs as root
05.25

4 environment files are committed

adequate

console/frontend/.env.production, docker/astronAgent/.env.with-rpa, docker/astronAgent/astronRPA/.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
console/frontend/.env.production — committed environment file
docker/astronAgent/.env.with-rpa — committed environment file
docker/astronAgent/astronRPA/.env — committed environment file
05.27

The build runs third-party code it does not pin

adequate

7 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; 3 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-push.yml:133 — action pinned by tag
.github/workflows/build-push.yml:136 — action pinned by tag
.github/workflows/ci.yml:152 — download piped into a shell
05.13

6 outbound calls can wait forever

adequate

6 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
core/common/service/oss/ifly_storage_gateway_service.py:64 — outbound call without a timeout
core/common/service/oss/s3_service.py:183 — outbound call without a timeout
core/workflow/engine/nodes/base_node.py:1200 — outbound call without a timeout
05.20

4 SQL statements built from strings

adequate

4 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.

05.22

9 deserialisers that can run code

adequate

9 places read data with a deserialiser that executes what it reads (pickle, `yaml.load` without a safe loader). That is safe on data the system wrote itself and remote code execution on anything else. Ask where each one's input comes from.

Evidence
core/common/service/cache/redis_cache.py:86 — code-executing deserialiser
core/common/service/cache/redis_cache.py:153 — code-executing deserialiser
core/common/service/cache/redis_cache.py:190 — code-executing deserialiser
05.26

83 locked dependencies have published vulnerabilities, 66 high or critical in what ships

adequate

83 of 1,397 locked dependencies match 192 published advisories in OSV, 86 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
core/workflow/uv.lock — certifi 2024.7.4: GHSA-6q48-6c7w-9gw2 (critical), fixed in 2026.10.19
core/common/poetry.lock — certifi 2024.8.30: GHSA-6q48-6c7w-9gw2 (critical), fixed in 2026.10.19
core/agent/uv.lock — certifi 2025.1.31: GHSA-6q48-6c7w-9gw2 (critical), fixed in 2026.10.19
05.45

2 workload settings give a container the host

adequate

2 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
console/frontend/deployment.yml:17 — privileged or host namespace
docker/ragflow/docker-compose-base.yml:111 — privileged or host namespace
05.65

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

adequate

The newest was published on 2026-10-11. 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 1 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 1.1.2. Answer lives in: the release you run, set against each fixed version.

Evidence
2026-10-11: CVE-2026-108864 (low): iFlytek Astron Agent through 1.1.2 contains an insecure direct object reference vulnerability that allows authenticated applications to resume other application
2026-09-07: GHSA-mh3w-4q3f-2fg5 (critical), fixed in v1.1.2: Unsandboxed code-node leads to cross-tenant RCE

Observability & Operations

adequate
06.2

No structured logging library is in use

adequate

Nothing references a structured logging library. Logs that are written to be read by a person are not searchable by a machine, which means incident response here is proportional to how well someone remembers the codebase rather than to what the tooling can find.

Evidence
searched 3,496 files for 10 structured logging libraries; 0 matched
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

10 of 28 deployed images carry no version, or `latest`

adequate

10 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
helm/astron-agent/values.yaml:224 — image without a version
helm/astron-agent/values.yaml:249 — image without a version
helm/astron-agent/values.yaml:273 — image without a version
06.14

Readiness and liveness probes are identical in 1 of 9 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
helm/astron-agent/templates/auth/casdoor-deployment.yaml — same probe endpoint twice
06.33

2 of 10 deployments run a single replica

adequate

2 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 iflytek/astron-agent: 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.