T09 · Insecure Skill Coding Practices
- Location
server.py:6- Finding
TLS Certificate Verification Disabled for Authenticated API Requests
- Content
View full analysis
- Remediation
View remediation
Security audit
Security checks for vulnerabilities and agentic risk
This skill does what it says, but it sends an infrastructure API token over HTTPS while deliberately disabling certificate verification.
Review before installing. Only run this with a least-privilege, read-only SDDC Manager token and restricted outbound network access. The TLS behavior should be fixed first by enabling certificate verification or configuring a trusted private CA bundle; otherwise the token and pre-check data could be exposed on an intercepted network path. Pin dependencies before production use.
server.py:6TLS Certificate Verification Disabled for Authenticated API Requests
requirements.txt:1Unpinned Third-Party Python Dependencies
Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.
}
try:
response = requests.get(url, headers=headers, verify=False)
response.raise_for_status()
data = response.json()
The skill requires environment-provided secrets and network access to a remote SDDC Manager API, but it does not declare any explicit tool scope such as permissions or allowed-tools. This creates a trust and containment gap: an agent or operator may enable the skill without a clear policy boundary around secret access and outbound connectivity, increasing the chance of over-privileged execution or unintended data exposure.
This code sends a bearer token from an environment variable to a remote API via an HTTP request. Although the docstring explains the tool's purpose, it does not disclose that it will use credential material from the environment and transmit it over the network, and there is no user-facing prompt or logging notice before doing so.
The code disables TLS certificate validation with verify=False and suppresses the related warnings globally. This allows an attacker on the network path to impersonate the SDDC Manager endpoint, intercept the bearer token, and tamper with pre-check results or remediation output.
}
try:
response = requests.get(url, headers=headers, verify=False)
response.raise_for_status()
data = response.json()
The dependency mcp is unpinned, so installs may resolve to different versions over time, including vulnerable or breaking releases. In a network-facing MCP server context, this is more dangerous because the package directly underpins protocol handling and may expose the skill to newly introduced or known SDK vulnerabilities without any change to this repository.
mcp
requests
mcp has multiple known advisories, and because no version is pinned, it is impossible to determine whether deployments will install an affected release. Given this skill is itself an MCP server, any vulnerable mcp version is especially relevant because flaws in protocol parsing, server validation, or access control could be directly reachable by clients.
The dependency requests is also unpinned, making builds non-reproducible and allowing future installs to pull versions with known defects or behavior changes. Because this skill communicates with VCF SDDC Manager over HTTP(S), weaknesses in the HTTP client library could affect credential handling, TLS verification behavior, or request safety.
mcp
requests
requests has numerous historical advisories, and without version pinning there is no assurance that installations avoid affected releases. Because this skill likely uses requests to connect to management infrastructure, a vulnerable HTTP client could contribute to credential exposure, SSRF-adjacent issues, or weakened transport security depending on the installed version and usage.