registry / @mozartec/mos-cli / 0.5.0

@mozartec/mos-cli@0.5.0

The mos CLI: serve any vault's board and wiki locally (`mos serve`), scaffold a new vault (`mos init`), and validate one (`mos validate`). Read-only over existing vaults; ships the built web app bundled with the vault endpoints in one Node process.

AI Security Review

scanned 10h ago · by lpm-firewall-ai

Review flagged AI-agent configuration or capability changes. This remains warn-only unless evidence shows foreign-agent hijack through preinstall/install/postinstall, hidden persistence, exfiltration, remote code execution, or other concrete malicious behavior.

Static reason

No blocking static signals were detected.

Trigger

User runs `mos init [dir]` or `mos serve`.

Impact

`init` can add project-local agent instruction files to its selected target directory.

Mechanism

User-invoked vault scaffolding and localhost read-only server.

Rationale

The package is not concretely malicious, but its explicit initialization command writes AI-agent instruction files. Under the stated policy this is a non-blocking warn-level capability.

Evidence

package.jsonbin/mos.mjsdist/main.jsREADME.mdAGENTS.md.mos/AGENTS.md.mos/config.jsonboard/T-001-explore-the-board.md

Network endpoints1

127.0.0.1:<port>

Decision evidence

public snapshot

AI called this Suspicious at 91.0% confidence as Dangerous Capability with false-positive risk.

@mozartec/mos-cli@0.5.0

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Source & flagged code