ClawHub
Back to skill

Security audit

Agent Archive

Security checks for vulnerabilities and agentic risk

Overview

This skill does what it says: it helps export agent conversations locally, but users should treat the resulting files as sensitive records.

Install this only if you want durable local archives of agent sessions. Review exported files before sharing or syncing them, because they may contain prompts, file paths, tool outputs, credentials pasted into chat, or other sensitive project details; prefer compact exports for routine use and store full/replay/jsonl exports in a controlled location.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

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 (4)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill explicitly supports exporting full conversations, tool call persistence, session metadata, and multi-format archives to disk, which can capture secrets, personal data, access tokens, file contents, and sensitive tool outputs without any warning, consent checkpoint, or redaction guidance. In an agent environment, archived conversations often contain exactly the highest-value data, so silent persistence materially increases confidentiality and retention risk.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

Replay and jsonl/dataset-oriented export modes encourage preserving chronological execution, tool calls, arguments, results, and assistant summaries in formats suitable for reuse, benchmarking, or training. Without safeguards, this can turn sensitive operational transcripts into durable datasets that are easier to redistribute, ingest into RAG/training systems, or replay outside the original trust boundary.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The script exports the full conversation payload, including message content and tool results, directly to a local file without any built-in confirmation, warning, redaction, or consent mechanism. In the context of an archiving skill, this is especially sensitive because conversations may contain secrets, personal data, credentials, or tool output from other systems, so a user can unintentionally persist sensitive information to disk where it may be exposed later.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

The skill provides default save paths for local archives, which normalizes automatic creation of persistent files containing conversation records. Even if the paths are local, users may not realize sensitive transcripts will remain on disk, be synced by backup tools, or be readable by other processes or users depending on system configuration.

Content

No source excerpt is available for this finding.