claude-api
Warn
Audited by Snyk on Jun 26, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.80). The skill explicitly instructs the agent to WebFetch external SDK/docs at runtime (see the referenced docs URL example https://github.com/anthropics/anthropic-sdk-python/blob/main/examples/memory/basic.py and MCP endpoints like https://api.githubcopilot.com/mcp/), which would be fetched during runtime and can directly control generated code, prompts, or agent tool integration—constituting a runtime external dependency that steers agent behavior.
Issues (1)
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
Audit Metadata