registry / @macintacos/caret / 0.2.0

@macintacos/caret@0.2.0

A Claude Code plugin that replaces the terminal plan-approval prompt with a local web UI for inline review, annotation, and approval.

AI Security Review

scanned 5h ago · by lpm-firewall-ai

LPM treats this as warn-only first-party agent extension lifecycle risk. This is a first-party Claude/OpenCode plan-review extension. Its daemon serves review data locally and its OpenCode installation path is explicitly user-invoked; no unconsented npm lifecycle mutation or remote exfiltration was confirmed.

Static reason

No blocking static signals were detected.

Trigger

Claude hook invocation or explicit `caret install-opencode` command.

Impact

Can gate plan approval and, when explicitly installed, add package-owned OpenCode integration files.

Mechanism

local plan-review daemon with optional OpenCode extension installation

Rationale

Source inspection found a legitimate local plan-review plugin with explicit agent-extension installation behavior, but no install-time malicious execution, credential theft, remote payload delivery, or foreign broad control-surface mutation. Warn under the extension-lifecycle policy rather than block.

Evidence

package.jsonbin/carethooks/hooks.jsondist/cli.jsopencode/caret.plugin.tscommands/discovery.mdcommands/demo.mdcommands/debug.md

Network endpoints2

localhost127.0.0.1

@macintacos/caret@0.2.0

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Source & flagged code