AOI Sandbox Shield (Lite)
Analysis
This appears to be a local snapshot and JSON-validation helper with no network or apply actions, though it can leave local copies of OpenClaw config and memory files.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.
Source: unknown; Homepage: none
The registry metadata does not provide an independent source repository or homepage for provenance review. The bundled package has no dependencies or install script, so this is a notice rather than a concern.
Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.
const snapRoot = path.join(workspace, '.sandbox_snapshots_lite'); ... path.join(workspace, 'CURRENT_STATE.md'), path.join(workspace, 'MEMORY.md') ... fs.copyFileSync(p, path.join(dir, base));
The snapshot command copies local agent state and memory/context files into a persistent snapshot directory. This is aligned with the snapshot purpose, but it duplicates potentially private local context.
