84%
Total Score
healthy
Active releases and a broad contributor base outweigh workflow permission and repository-linkage concerns.
The linked repository is the Vercel AI monorepository, so its name does not match the package and its README does not mention @ai-sdk/anthropic. The monorepo structure makes the name mismatch understandable, but the lack of a README mention leaves some package-to-repository linkage ambiguity.
All 13 workflows were analyzed successfully and all 78 action references are pinned, but the audit found high-confidence github-app findings, a high-confidence template-injection finding, and one untrusted checkout in a workflow_run workflow; two workflows also grant top-level write access. These are meaningful workflow-hygiene concerns, though no script injection was reported and most permissions are read-only.
We didn't find any vulnerabilities for this package.
| Dependency | Last Release | Score |
|---|---|---|
@ai-sdk/provider Version 4.0.26 | — | — |
@ai-sdk/provider-utils Version 5.0.58 | — | — |
Connect your repositories to instantly see whether vulnerable or malicious packages exist in your codebase.
Free. No credit card required.

I consent to receiving marketing communications based on Aikido’s Privacy Policy.
SOC 2Compliant
ISO 27001Compliant
ISO 42001Compliant© All Intel data is openly available and commercially licensed.