i agree. just using right now the coding agent to do so, coding, small files, functions, lines of code with mandatory rules and guardrails, scope, and context.
i have similar issue, working by yourself need some times adverse ideas, opposition, feedback. even the LLM tend to be complacent. i found myself (soudn crazy or not) making JIRA projects jsut to keep tracking of progress or ideas.
Thanks for the link. interesting to know about the work of go plus team, what i am seeing, GoPlus AgentGuard is focused on skill/plugin
security — static analysis of code before it runs, Web3 exploits,
trust registries. Mine is focused on filesystem watching during
live sessions — what files the agent touches, Telegram rollback,
persistent daemon. Different layers of the same problem. The name
collision is unfortunate but the tools are complementary.