datachain-ai/datachain
The Context Layer for unstructured data: typed, versioned datasets over S3, GCS, Azure
- Repository: https://github.com/datachain-ai/datachain
- Commit:
8b7fe3202ba87f473782b06ac5e184d23d674921 - Generated: 2026-11-14T00:06:01+00:00
- Scorer 0.1.0, scoring prov-4+qual-2+sec-2, report schema v2
Scores
| Pillar | Score | Signal | Confidence | Coverage |
|---|---|---|---|---|
| AI provenance (estimated AI-generated share) | 9 | explicit | 95% | 100% |
| Code quality | 76 | adequate | 90% | 100% |
| Security | 63 | weak | 85% | 100% |
AI provenance
Explicit AI attribution on 9% of first-party code added; an estimated 9% including velocity signals.
- Explicitly attributed share (lower bound): 9.3%
- Unattributed share in AI-consistent velocity bursts: 1.1%
- Estimated AI-generated share: 9.3%
| AI tool | Attributed commits | Lines added |
|---|---|---|
| Claude Code | 87 | 15,643 |
| GitHub Copilot coding agent | 21 | 2,873 |
| Cursor agent | 3 | 1,039 |
| OpenAI Codex | 1 | 195 |
- AI tooling configured in repo: claude-code
Current source lines last changed by AI-attributed commits: 6,165 of 60,658.
| Findings per 1,000 lines | AI-attributed | Other | Ratio |
|---|---|---|---|
| Quality | 2.595 | 2.147 | 1.21x |
| Security | 0.649 | 0.312 | 2.08x |
- Stylometry index: 0.22 per 1,000 lines (narrated steps 11, placeholders 1, assistant phrasing 0, emoji 1). Reported as a signal, not scored.
| Month | Lines added | Attributed to AI |
|---|---|---|
| 2025-12 | 5,326 | 4% |
| 2026-01 | 2,640 | 0% |
| 2026-02 | 6,909 | 33% |
| 2026-03 | 10,403 | 7% |
| 2026-04 | 10,375 | 0% |
| 2026-05 | 4,428 | 3% |
| 2026-06 | 5,246 | 4% |
| 2026-07 | 8,021 | 17% |
| 2026-08 | 3,074 | 50% |
| 2026-09 | 1,517 | 39% |
| 2026-10 | 15,329 | 32% |
| 2026-11 | 15,811 | 42% |
Evidence
- Claude Code: PR description attribution (8b7fe3202ba8)
- Claude Code: trailer 'Co-Authored-By: Claude' (ae3ce429b3f8)
- Claude Code: PR description attribution (5de86b14002e)
- Claude Code: PR description attribution (3e034bbdb4f9)
- Claude Code: trailer 'Co-Authored-By: Claude' (c726a3ce002d)
- Claude Code: trailer 'Co-Authored-By: Claude' (42f7aef8fd27)
- Claude Code: trailer 'Co-authored-by: Claude' (fcbf2bfb5aae)
- Claude Code: trailer 'Co-Authored-By: Claude' (eac393917eec)
- Claude Code: trailer 'Co-Authored-By: Claude' (fdd56ac26014)
- Claude Code: trailer 'Co-authored-by: Claude' (a50eafa67bb8)
- Claude Code: PR description attribution (290720176d80)
- Claude Code: trailer 'Co-Authored-By: Claude' (7661288e7a0a)
- Claude Code: PR description attribution (5ddebc3deb65)
- Claude Code: PR description attribution (aa5ef5598ed3)
- Claude Code: PR description attribution (37de2f7c70a6)
AI provenance describes how code was produced; it is not a quality judgment. Unless attribution is explicit, these are signals consistent with AI generation, not proof. Squash/rebase workflows, imported code, and rewritten history reduce accuracy.
Code quality
Code quality is adequate: test-to-source ratio 1.42, CI runs tests; 14% of function code in complex or long functions; 5% duplicated lines. Weakest area: lint.
| Area | Score |
|---|---|
| Complexity (lizard) | 66 |
| Duplication | 75 |
| Code smells (Opengrep) | 55 |
| Tests and CI | 100 |
| Documentation | 80 |
| Type safety | 60 |
| Process (Scorecard review/maintenance) | n/a |
- Functions: 2,975; mean complexity 3.3, 90th percentile 8; 53 complex and 15 long functions
- Duplicated lines: 4.8% of 44,631 compared
- Tests: 85,999 test lines for 60,658 source lines (ratio 1.42); CI test commands: pytest, nox; 84% of recent source commits also change tests
- Docs missing or thin: changelog
Most complex functions:
- init in src/datachain/lib/settings.py: complexity 37, 143 lines
- group_by in src/datachain/lib/dc/datachain.py: complexity 33, 134 lines
- _compare in src/datachain/diff/init.py: complexity 31, 106 lines
- pull_dataset in src/datachain/catalog/catalog.py: complexity 30, 247 lines
- type_to_str in src/datachain/lib/utils.py: complexity 28, 114 lines
Security
Security posture is weak: 8 high-severity SAST findings; 37 high-severity workflow issues. Weakest area: ci actions.
| Area | Score |
|---|---|
| Secrets (gitleaks, tree + history) | 77 |
| Static analysis (Opengrep) | 94 |
| Vulnerable dependencies (OSV) | n/a |
| CI workflows (zizmor) | 0 |
| Posture (policy, updates, signing, lockfiles) | 60 |
| OpenSSF Scorecard | n/a |
- Posture checks passed: dependency updates, signed commits, lockfiles; missing: security policy
Top findings
| Severity | Finding | Location |
|---|---|---|
| high | opengrep/python.lang.security.audit.dangerous-system-call-audit: Found dynamic content used in a system call. This is dangerous if external data can reach this function call because it allows a malicious actor to execute commands. Use the 'subprocess' module instead, which is easier to use without accidentally exposing a command injection vulnerability. [confidence low] |
docs/api_hooks.py:42 |
| high | opengrep/python.lang.security.use-defused-xml: The Python documentation recommends using defusedxml instead of xml because the native Python xml library is vulnerable to XML External Entity (XXE) attacks. These attacks can leak confidential data and "XML bombs" can cause denial of service. [confidence low] |
src/datachain/client/azure.py:8 |
| high | opengrep/python.lang.security.audit.dangerous-asyncio-create-exec-audit: Detected 'create_subprocess_exec' function without a static string. If this data can be controlled by a malicious actor, it may be an instance of command injection. Audit the use of this call to ensure it is not controllable by an external resource. You may consider using 'shlex.escape()'. [confidence low] |
src/datachain/client/listing_worker.py:83 |
| high | opengrep/python.lang.security.use-defused-xml: The Python documentation recommends using defusedxml instead of xml because the native Python xml library is vulnerable to XML External Entity (XXE) attacks. These attacks can leak confidential data and "XML bombs" can cause denial of service. [confidence low] |
src/datachain/client/s3.py:7 |
| high | opengrep/python.lang.security.audit.dangerous-subprocess-use-audit: Detected subprocess function 'Popen' without a static string. If this data can be controlled by a malicious actor, it may be an instance of command injection. Audit the use of this call to ensure it is not controllable by an external resource. You may consider using 'shlex.escape()'. [confidence low] |
src/datachain/lib/video.py:616 |
| high | opengrep/python.lang.security.audit.dangerous-subprocess-use-audit: Detected subprocess function 'Popen' without a static string. If this data can be controlled by a malicious actor, it may be an instance of command injection. Audit the use of this call to ensure it is not controllable by an external resource. You may consider using 'shlex.escape()'. [confidence low] |
src/datachain/query/dataset.py:1010 |
| high | opengrep/python.lang.security.audit.dangerous-subprocess-use-audit: Detected subprocess function 'run' without a static string. If this data can be controlled by a malicious actor, it may be an instance of command injection. Audit the use of this call to ensure it is not controllable by an external resource. You may consider using 'shlex.escape()'. [confidence low] |
src/datachain/sql/sqlite/base.py:779 |
| high | opengrep/python.sqlalchemy.security.audit.avoid-sqlalchemy-text: sqlalchemy.text passes the constructed SQL statement to the database mostly unchanged. This means that the usual SQL injection protections are not applied and this function is vulnerable to SQL injection if user input can reach here. Use normal SQLAlchemy operators (such as or_, and_, etc.) to const [confidence medium] |
src/datachain/testing.py:25 |
| high | zizmor/unpinned-uses: unpinned action reference [confidence high] |
.github/actions/install-ffmpeg/action.yml:17 |
| high | zizmor/unpinned-uses: unpinned action reference [confidence high] |
.github/actions/install-ffmpeg/action.yml:37 |
| high | zizmor/unpinned-uses: unpinned action reference [confidence high] |
.github/workflows/benchmarks.yml:21 |
| high | zizmor/unpinned-uses: unpinned action reference [confidence high] |
.github/workflows/benchmarks.yml:23 |
| high | zizmor/unpinned-uses: unpinned action reference [confidence high] |
.github/workflows/benchmarks.yml:28 |
| high | zizmor/unpinned-uses: unpinned action reference [confidence high] |
.github/workflows/codspeed.yml:20 |
| high | zizmor/unpinned-uses: unpinned action reference [confidence high] |
.github/workflows/codspeed.yml:25 |
| high | zizmor/unpinned-uses: unpinned action reference [confidence high] |
.github/workflows/codspeed.yml:30 |
| high | zizmor/unpinned-uses: unpinned action reference [confidence high] |
.github/workflows/codspeed.yml:40 |
| high | zizmor/unpinned-uses: unpinned action reference [confidence high] |
.github/workflows/codspeed.yml:49 |
| high | zizmor/unpinned-uses: unpinned action reference [confidence high] |
.github/workflows/codspeed.yml:54 |
| high | zizmor/unpinned-uses: unpinned action reference [confidence high] |
.github/workflows/codspeed.yml:59 |
| high | zizmor/unpinned-uses: unpinned action reference [confidence high] |
.github/workflows/codspeed.yml:72 |
| high | zizmor/unpinned-uses: unpinned action reference [confidence high] |
.github/workflows/release.yml:20 |
| high | zizmor/unpinned-uses: unpinned action reference [confidence high] |
.github/workflows/release.yml:25 |
| high | zizmor/unpinned-uses: unpinned action reference [confidence high] |
.github/workflows/release.yml:30 |
| high | zizmor/unpinned-uses: unpinned action reference [confidence high] |
.github/workflows/release.yml:39 |
Static signals only: the scorer never runs the project's code. Secret findings are pattern matches that were not verified, and secret values are never stored.
Repository profile
- Source lines: 60,658; test lines: 85,999
| Language | Lines |
|---|---|
| Python | 146,657 |