Back to skill

Security audit

huawei-cloud-cloudrobo-resource

Security checks for vulnerabilities and agentic risk

Overview

This cloud-resource query skill needs review because its connection guidance weakens protection against interception and falsified results.

Before connecting a real account, require CLOUDROBO_VERIFY_SSL=true and confirm that the installed client rejects untrusted certificates. Repair certificate trust problems instead of using the documented bypass. Use credentials restricted to the listed read operations and keep traffic logging disabled. All ten bundled files were inspected: the scanner's missing click.Choice reference is a code symbol, while client.py belongs to the external dependency. That dependency's runtime behavior remains outside this artifact review.

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
references/cli-installation-guide.md
Finding

Troubleshooting instructions disable TLS certificate verification

Content
View full analysis

Impact

An active network attacker can inspect resource-query traffic and forge quota, capacity, or pool-detail responses. Request HMAC signing does not authenticate the TLS peer or protect response confidentiality and integrity. This does not establish disclosure of the secret signing key.

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (6)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
76% confidence
Finding

The supplied entrypoint appears to launch CLI-command tests, a materially different purpose from the declared resource-visibility operations. No quota queries, pool filtering, or node-detail retrieval are visible. This finding is provisional: without the invoked script's contents, it is not possible to verify whether the tests actually implement the declared workflow or perform other actions.

Content

No source excerpt is available for this finding.

The troubleshooting instructions recommend disabling TLS certificate verification without warning about credential exposure.

High
Category
Output Handling
Confidence
98% confidence
Finding

The troubleshooting advice disables server certificate verification for authenticated cloud requests, allowing a network attacker to impersonate the endpoint and inspect or alter resource-query traffic. The 'debug only' qualifier does not prevent exposure when debugging uses real credentials and cloud data; HMAC request signing does not authenticate the server or protect response confidentiality.

Content

No source excerpt is available for this finding.

Unsafe Defaults

Medium
Category
Tool Misuse
Confidence
98% confidence
Finding

The guide recommends disabling verification and separately documents its default as false, presenting an insecure connection posture as normal. If the client follows this documented default, even read-only queries can disclose infrastructure information or receive attacker-controlled results; the supplied packet does not independently verify the client's actual default.

Content

Scanner excerpt · references/cli-installation-guide.md (reported line 102)May include surrounding context.

md
| `HTTP 401/403` | Check AK/SK credentials in environment or config |
| `HTTP 403 ABAC` | Quota list and pool list require ABAC permission; check IAM policies |
| `HTTP 404` | Pool ID not found; verify with `list-pools` first |
| SSL verification errors | Set `CLOUDROBO_VERIFY_SSL=false` (debug only) |

## Environment Variables 环境变量

Unsafe Defaults

Medium
Category
Tool Misuse
Confidence
87% confidence
Finding

This passage is not itself a default setting, but it endorses disabling certificate verification during debugging without restricting that debugging to isolated systems and non-sensitive test credentials. Enabling verification in production does not protect authenticated debugging sessions from endpoint impersonation and disclosure or manipulation of cloud resource data.

Content

Scanner excerpt · references/iam-policies.md (reported line 86)May include surrounding context.

md
- **Signing mechanism** — APIG HMAC-SHA256 signs each request with a timestamp to prevent
  replay attacks
- **Proxy support** — Optional HTTP/HTTPS proxy can be configured for network isolation
- **SSL verification** — Can be disabled for debugging (`CLOUDROBO_VERIFY_SSL=false`) but
  should be enabled in production
- **Traffic logging** — `CLOUDROBO_LOG_TRAFFIC=true` enables request/response logging for
  debugging; disable in production to avoid credential leakage in logs

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
85% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · scripts/test-cli-commands.sh (reported line 47)May include surrounding context.

sh
echo "Credentials found: AK=${ak:0:8}..."
}

# Read-only test cases (auto-executed)
test_list_quotas() {
    echo "=== TC-01: list-quotas ==="
    if [ "$EXECUTOR" = "cli" ]; then

AC-23 requires bilingual documentation headings, including Chinese, without a language choice or stated justification.

Low
Category
Output Handling
Confidence
88% confidence
Finding

The acceptance criterion mandates bilingual section headers and verifies specific Chinese headings. This imposes a natural-language requirement on generated documentation without offering user opt-in or explaining a locale-specific need.

Content

No source excerpt is available for this finding.