Dual-provider LLM gateway for the Milpa PHP framework: OpenAI + Anthropic chat-completions client with tool-call format translation, an agentic tool-use loop, and a ToolRegistry facade for LLM/MCP exposure.
70%
Total Score
caution
Active releases and commits support adoption, but one contributor and unpinned CI actions add maintenance and build risks.
One contributor made all 89 commits in the last three months, creating a clear single-person continuity risk. Organization ownership provides some handoff capacity but does not remove the lack of an independently active contributor.
The repository recorded 89 commits in three months, showing active development, but all were made by one active maintainer.
Composer build tooling is present, but no security-scanning tools were detected. The missing scanning is a modest transparency and maintenance gap rather than evidence of an unsafe release.
v0.42.1 is not marked prerelease and recent releases have no prerelease share, but the 0.x major version means the API is not yet at major-version stability.
All four workflows were analyzed with no untrusted checkout, script injection, or auditor findings. However, all 8 action references are unpinned and three workflows grant top-level write permissions, creating reproducibility and least-privilege concerns without a dangerous trigger-and-sink combination.
We didn't find any vulnerabilities for this package.
| Dependency | Last Release | Score |
|---|---|---|
psr/log Version ^3 | — | — |
milpa/core Version >=0.6.2 <1.0 | — | — |
guzzlehttp/psr7 Version ^2.7 | — | — |
psr/http-client Version ^1.0 | — | — |
psr/http-factory Version ^1.1 | — | — |
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.