Back to skill

Security audit

File Exchange via Qiniu Kodo

Security checks for vulnerabilities and agentic risk

Overview

This file-transfer skill appears purpose-aligned but needs review because it uses cloud credentials, an unverified external tool, and third-party uploads without clear per-file user approval.

Review this before installing. Use it only if you intentionally want OpenClaw to transfer files through your Qiniu account, provide narrowly scoped and rotated credentials, verify qshell through a trusted channel, and confirm each file before upload.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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
Findings (2)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:5
Finding

Qshell Executable Downloaded Without Integrity Verification

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:14
Finding

Qiniu Secret Key Exposed Through Command-Line Arguments

Content
View full analysis
``` ```text QINIU_ACCESS_KEY QINIU_SECRET_KEY openclaw ``` The instructions require the access key and secret key to be obtained from the corresponding environment variables and substituted into the `qshell account` command, with `openclaw` used as the account name. ### Technical Analysis The authentication procedure places the Qiniu SecretKey directly in a command-line argument. Environment-variable storage does not prevent exposure once the value is expanded into the process argument vector. Depending on the operating system, process isolation, shell configuration, and execution framework, command arguments may be available through process-inspection interfaces, debugging or telemetry systems, shell tracing, audit logs, error reports, or agent tool-call logs. A local observer or logging service with access to these records could recover the credential. The risk is conditional on the surrounding runtime's process and logging protections, but passing a long-lived secret as a command argument creates an avoidable credential-exposure channel. ### Attack Path 1. The agent reads `QINIU_SECRET_KEY` and `QINIU_ACCESS_KEY` from its environment. 2. The values are expanded into the arguments of the `qshell account` process. 3. A local process observer, diagnostic collector, shell-tracing facility, or command-logging layer records or reads the process arguments. 4. An attacker with access to that process information or retained log data extracts the AccessKey and SecretKey. 5. The attacker authenticates to Qiniu using the exposed credentials. 6. The attacker performs storage operations permitted by the associated Qiniu account and credential policy. ### Impact Assessment Crede ...[truncated 643 chars]
Remediation
View remediation
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)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
81% confidence
Finding

The document is written as a hard requirement in Chinese and line L03 states an unconditional operational requirement for OpenClaw without indicating language flexibility or user opt-in. Under the stated policy, forcing a specific language or locale without choice can be a natural-language policy violation.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill instructs the agent to pull cloud AccessKey and SecretKey from environment variables and use them directly for authentication, but provides no safeguards on secret handling, scope minimization, logging hygiene, or user disclosure. In an agent setting, this creates a real risk of credential misuse or accidental exposure, especially because the skill normalizes automatic access to sensitive secrets as part of routine file transfer behavior.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The skill directs the agent to upload local files to a remote Qiniu bucket and return a cloud URI, but does not warn the user that files will leave the local system or require confirmation before exfiltration. In a general-purpose agent context, this is dangerous because it can silently transmit sensitive workspace data to third-party cloud storage under the skill author's preferred channel.

Content

No source excerpt is available for this finding.