agency-lang@0.17.1
The Agency language
AI Security Review
scanned 2mo ago · by lpm-firewall-aiThe flagged primitives implement explicit CLI features and include relevant containment or validation. The manifest has no lifecycle hook that would activate them during installation.
Decision evidence
public snapshot- The manifest has no install lifecycle hook; it exposes a CLI entrypoint.
- Child execution runs compiled user input with the shell disabled.
- Workspace writeback verifies each source hash before changing it.
- Subprocess files are confined to a package temporary directory.
- OAuth requires HTTPS except for local development callbacks.
- Agent setup and scheduling are explicit CLI subcommands.
- No install-time, import-time, or self-dependency attack chain was found.
- No secret harvesting or unconsented network exfiltration was found.
- Persistence-related behavior is reachable through explicit scheduling commands only.
Source & flagged code
6 flaggedPackage source references child process execution.
dist/lib/optimize/workspace.jsView on unpkg · L6Findings
Affected versions and remediation
This report applies to agency-lang@0.17.1.
See all versions of agency-lang for other recorded verdicts.
- Review the evidence and your use of agency-lang@0.17.1 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.