Back to skill

Security audit

阿里云证书自动续期(免 AK)

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill suspicious because scanners report broad cloud permissions and an OIDC trust rule that could let unrelated repositories assume the role; the review is based on incomplete coverage.

Consider the reported cloud access and repository trust risks before installing. The scanners report high-impact permissions and a trust rule that may allow unrelated repositories to use them. Skillspector coverage was partial: it scanned 75% of components, one file was partly inspected, a static analyzer was degraded, and the image was outside that analyzer's scope. This verdict is a review warning, not a claim that the author acted maliciously.

Vulnerability Patterns
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
  • 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 (2)

T05 · Unauthorized Access and Privilege Escalation

Error
Location
scripts/setup_oidc.py
Finding

Prefix wildcards let unrelated GitHub repositories assume the cloud role

Content
View full analysis

Impact

An unrelated repository can assume the role and exercise its attached permissions. Under the default policy, that includes account-wide OSS, Function Compute, DNS, and certificate-management operations.

T05 · Unauthorized Access and Privilege Escalation

Warning
Location
scripts/setup_oidc.py
Finding

Certificate automation receives unrestricted service permissions across the account

Content
View full analysis

Impact

The role can read or delete unrelated OSS data, alter unrelated DNS records, modify Function Compute resources, and manipulate certificates beyond the renewal task. This materially amplifies the repository-trust vulnerability.

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (7)

Tainted flow: 'req' from os.environ.get (line 33, credential/environment) → urllib.request.urlopen (network output)

Critical
Category
Data Flow
Confidence
90% confidence
Finding

Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.

Content

Scanner excerpt · scripts/oidc_sts_env.py (reported line 35)May include surrounding context.

python
url = os.environ['ACTIONS_ID_TOKEN_REQUEST_URL'] + '&audience=' + audience
    req = urllib.request.Request(url, headers={
        'Authorization': 'Bearer ' + os.environ['ACTIONS_ID_TOKEN_REQUEST_TOKEN']})
    with urllib.request.urlopen(req, timeout=30) as r:
        return json.load(r)['value']

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

Checking certificate expiration could support the declared workflow, but this chunk consists entirely of read-only inspection. It contains no certificate issuance or renewal, GitHub Actions configuration, RAM OIDC authentication, or deployment to Alibaba Cloud services. Thus, the supplied chunk does not implement the materially broader declared purpose. No credential access or infrastructure modification is shown.

Content

No source excerpt is available for this finding.

The setup instructs granting full access across several cloud services without warning about the resulting account-wide impact.

High
Category
Output Handling
Confidence
99% confidence
Finding

The instructions prescribe wildcard actions across DNS, storage, functions, certificate management, and CDN rather than the specific operations required for renewal. A compromised workflow or assumed-role session could consequently modify or delete unrelated infrastructure, access data, or redirect traffic; temporary credentials limit duration but not this authorization scope.

Content

No source excerpt is available for this finding.

The default OIDC role receives service-wide permissions beyond certificate renewal and distribution.

High
Category
Excessive Agency
Confidence
99% confidence
Finding

The generated policy combines service-wide default actions with Resource ['*'], granting account-wide access across OSS, FC, DNS, and certificate management. Compromise of an authorized GitHub workflow therefore permits substantially more than certificate renewal, including potentially destructive operations and access to unrelated resources.

Content

No source excerpt is available for this finding.

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: env, file_write, network.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

Troubleshooting instructions expand into account-wide private-repository discovery and extraction of cached GitHub credentials.

Medium
Category
Excessive Agency
Confidence
91% confidence
Finding

The troubleshooting guidance expands a certificate-renewal task into account-wide private-repository enumeration and retrieval of an existing GitHub credential without an explicit authorization checkpoint. Although the credential-helper interface is legitimate and no attacker-controlled destination is shown, this can expose unrelated repository metadata and place a broadly privileged token into shell and process contexts.

Content

No source excerpt is available for this finding.

The troubleshooting workflow instructs disabling local proxy settings without warning about bypassing network controls.

Medium
Category
Output Handling
Confidence
91% confidence
Finding

The guidance instructs users to unset local proxy variables before making authenticated GitHub API requests without checking whether that proxy is an organizational security control. Where the proxy provides mandatory filtering or auditing, following this instruction can route credential-bearing traffic around those controls, although HTTPS itself is not disabled.

Content

No source excerpt is available for this finding.