agency-lang@0.19.0
The Agency language
AI Security Review
scanned 2mo ago · by lpm-firewall-aiNo concrete malicious install hook, stealth persistence, or unauthorized data exfiltration was found. Because the package is an agent-oriented language runtime with broad user-code execution and the supplied static analysis reported incomplete coverage, it should be warned on rather than marked clean.
Decision evidence
public snapshot- The package exposes a CLI that compiles and launches supplied Agency programs in a child Node process.
- Its runtime materializes compiled program code under the package directory before forking a subprocess.
- OAuth support can open a browser, accept a local callback, and exchange a user-configured authorization code for tokens.
- package.json has no preinstall, install, or postinstall lifecycle hook.
- The reviewed process launches use argument arrays and explicitly disable the shell where applicable.
- The OAuth flow restricts non-local authorization and token endpoints to HTTPS and validates callback state.
- The local model readiness request targets only 127.0.0.1.
Source & flagged code
7 flaggedPackage source references child process execution.
dist/lib/optimize/workspace.jsView on unpkg · L6Findings
Affected versions and remediation
This report applies to agency-lang@0.19.0.
See all versions of agency-lang for other recorded verdicts.
- Review the evidence and your use of agency-lang@0.19.0 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.