Caveat verdict
claude-code-agent-sdk
Documentation skill for the Claude Code Agent SDK; provides reference documentation with no execution capabilities.
⚠ Flagged for review — coarse, uncorroborated signal, not a confirmed exploit. Review the config yourself before installing.
Automated static analysis — not a human review. Caveat flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.
Findings (5)
Pipe to python — executes piped content as Python code
references/file-checkpointing.md · prose · downgraded · | Python
Pipe to bash — executes piped content as shell commands
references/permissions.md · prose · downgraded · | Bash
References agent configuration files
references/modifying-system-prompts.md · code · CLAUDE.md
Accesses sensitive environment variables
references/mcp.md · code · process.env.API_TOKEN
Opens WebSocket connection
references/hosting.md · prose · downgraded · WebSocket
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class A). Final tier capped at Caution — cannot be lifted by any downgrade, example-payload opt-in, or allowlist.
Permissions & capabilities
No declared permissions — minimal attack surface.
network_in Is this flag fair?
Thanks — recorded.