Back to skill

Security audit

asclaude-compact

Security checks for vulnerabilities and agentic risk

Overview

This is a local OpenClaw conversation archiving and compaction skill whose sensitive file access is disclosed and fits its purpose, but users should treat its saved archives as private data.

Install only if you are comfortable with local copies and summaries of OpenClaw conversations being written under ~/.openclaw/workspace. Avoid running it on sessions containing secrets unless you secure or delete the generated archive, summary, and backup files afterward.

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The documentation promotes archiving, summarization, and deletion of conversation content but does not warn users that these actions modify stored conversation state and can permanently remove useful detail. In a memory-management skill, silent data reduction can cause loss of context, auditability, or user expectations around retention, especially if a user assumes the process is reversible or non-destructive.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The script reads complete local conversation histories and persists full archived copies plus derived summaries without any consent prompt, warning, or minimization. Those session files may contain secrets, credentials, personal data, or proprietary prompts, and silently duplicating them increases exposure if the workspace is shared, backed up, synced, or later accessed by other tools.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The summary file includes previews of recent messages and writes them in plaintext to the workspace memory directory without warning the user. Even truncated previews can leak sensitive prompts, personal information, API keys, internal URLs, or other confidential data, especially in environments where workspace files are searchable, synchronized, or visible to collaborators.

Static analysis

No suspicious patterns detected.