Back to skill

Security audit

flow-monitor

Security checks for vulnerabilities and agentic risk

Overview

This skill is a local trace summarizer that reads a nearby execution log and reports timing and size metadata without network access or persistence.

Before installing, understand that this skill reads local execution logs to summarize recent tool calls. Use it only in workspaces where reading claw_execution.log is acceptable, because that log may contain sensitive tool inputs even though this script only outputs size and estimate metadata.

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
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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 (2)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The script extracts with input content from the log and outputs derived metadata about it, which can expose user-provided or system-generated data present in the log. There is no comment, prompt, or user-facing notice warning that logged tool inputs will be read and surfaced by this script.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
94% confidence
Finding

Several code comments are written in Chinese with no indication that language choice is optional or context-specific. Under the stated policy, forcing a specific language without user opt-in is a natural-language locale violation.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.