Back to skill

Security audit

arkcli-connect

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill suspicious because its default synchronization is reported to remove third-party skills and redirect agent requests, while part of the reference coverage is incomplete.

Before installing, consider whether you want synchronization to remove other installed skills and change request routing. The report says these changes happen by default; review the behavior and the missing referenced path before relying on the scan as complete.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (1)

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md
Finding

Default installation expands into destructive changes without explicit consent

Content
View full analysis

Impact

Existing customized instructions can be lost, third-party capabilities removed, and future generation requests rerouted. The documented behavior is a concrete destructive scope expansion, not evidence of credential theft or a backdoor.

Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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
Findings (3)

Default synchronization also removes third-party skills to redirect agent routing.

Medium
Category
Excessive Agency
Confidence
97% confidence
Finding

Default installation deletes two explicitly named third-party skills to make generation requests route to ArkCLI, exceeding the user's ordinary authorization to install or synchronize ArkCLI skills. Exact-name matching and visible logging limit accidental scope but do not establish ownership or consent; automatic postinstall execution further increases the risk of unexpected loss of user-managed skills and changes to agent behavior.

Content

No source excerpt is available for this finding.

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
80% confidence
Finding

Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Content

Scanner excerpt · references/arkcli-connect.md (reported line 50)May include surrounding context.

落盘结构:

text
/path/to/repo/.claude/skills/arkcli-chat/SKILL.md
/path/to/repo/.claude/skills/arkcli-gen/SKILL.md
/path/to/repo/.claude/skills/.arkcli-managed-skills.json
...

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
80% confidence
Finding

Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Content

Scanner excerpt · references/arkcli-connect.md (reported line 51)May include surrounding context.

text
/path/to/repo/.claude/skills/arkcli-chat/SKILL.md
/path/to/repo/.claude/skills/arkcli-gen/SKILL.md
/path/to/repo/.claude/skills/.arkcli-managed-skills.json
...