Memory Read
Security checks across malware telemetry and agentic risk
Overview
This instruction-only memory retrieval skill appears purpose-aligned and bounded, but users should notice that it can read stored personal context or chat snippets and has a small metadata mismatch.
This skill is reasonable for a personal memory system, but review what your indexed memory areas contain before enabling it. Pay special attention to raw chat logs and any sensitive information that could be pulled into the agent's context.
VirusTotal
62/62 vendors flagged this skill as clean.
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.
Stored memories or chat excerpts could be surfaced to the agent and shape its responses.
The skill is designed to read persistent memory and, in a manual trigger case, raw cold chat snippets. This is purpose-aligned for a memory-reading skill and includes limits, but the content may be sensitive and can influence future agent context.
失忆后重建 | 小蜂日报/最近7天 ... 调取聊天 | 冷/ | 仅彧哥指令触发，手动，≤50KB
Install only if you want the agent to use these memory stores; keep sensitive secrets out of indexed memory and confirm before requesting cold chat retrieval.
The package identity/version information is inconsistent, which may make provenance tracking less clear.
The internal metadata does not match the registry metadata. Because there is no executable code or install mechanism, this is a provenance note rather than evidence of unsafe behavior.
Registry: Owner ID kn7fashbtj7p1hr48327nx3wes82xrbt, Version 1.0.4; _meta.json: "ownerId": "kn73vp5rarc3b14rc7wjcw8f8580t5d1", "version": "1.0.3"
Verify the publisher and intended version before relying on this skill, especially if future versions add code or external dependencies.