@hmharness/agent@0.14.24
hmharness agent execution layer: base tools, system prompt, sub-agent spawn, and the shared task runner that frontends (cli, web) drive.
AI security review
scanned · by lpm-firewall-aiThe manifest defines no install lifecycle hook, and the inspected behaviors are runtime features invoked through the agent or project APIs. The agent instruction file read is tied to the caller's working directory and is passed into prompt construction; no concrete malicious behavior was found.