Hippocampus Subagent Memory
Security checks across static analysis, malware telemetry, and agentic risk
Overview
This instruction-only skill is a coherent guide for keeping sub-agent memories separated and only merging approved outputs, with no code, install steps, or credential access shown.
This appears safe to install as an instruction-only memory-isolation aid. Before using it, make sure your workflow reviews what child-agent memory is imported back to the parent and avoids copying full transcripts or sensitive data unnecessarily.
Static analysis
No static analysis findings were reported for this release.
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
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.
If used, sub-agent memories may persist or be reused, so users should ensure only intended summaries or artifacts are imported into the parent memory.
The skill manages cross-context memory state, which can affect what future agents remember or reuse. The same text also provides containment by requiring namespacing and explicit approval before merge-back.
Let the child write only into its own namespace... Import back to parent only what is explicitly approved.
Keep sub-agent namespaces scoped, review any proposed merge-back content, and avoid importing full transcripts or untrusted instructions into parent memory.