Caveat verdict
multi-agent-brand-studio
Multi-agent social media operations setup skill that creates specialized agents with approval-gated publishing; agent_memory is expected for multi-agent coordination and the approval workflow explicitly prevents unauthorized publishing.
⚠ 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.
Permission integrity
agent_memory
Findings (7)
Pipe to python — executes piped content as Python code
assets/skills/qmd-setup/SKILL.md · code · | python3
Accesses OpenClaw config/secrets directly
SKILL.md · code · ~/.openclaw/openclaw.json
References agent memory files
SKILL.md · code · MEMORY.md
Instructs covert action — may act without user awareness
assets/shared/operations/communication-signals.md · prose · downgraded · silently
Uses exec() — may execute shell commands
references/agent-roles.md · prose · downgraded · exec (
References child_process — can spawn system processes
scripts/patch-config.js · prose · downgraded · child_process
Node http/https module — low-level network access
scripts/telegram-topics.js · prose · downgraded · require("https")
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class D). 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.
agent_memory Is this flag fair?
Thanks — recorded.