Hermes Adds Shell Hooks and /goal
alexcovo_eth · x · 2026-07-16
This post introduces two key capabilities in Hermes: shell hooks and /goal.
Shell hooks
Many users don't know they can use hooks to constrain behavior via code, such as:
- Blocking command execution
- Formatting files
- Injecting context
- Triggering workflows
- Auditing agent activity
It requires no Python plugins, just YAML + scripts.
/goal
/goal turns Hermes into an autonomous task executor:
- The user provides a task objective, such as building a FastAPI authentication service
- The system continues to progress without needing re-prompting every turn
- An internal judge determines if the result is complete / continue / pending
- By default, it can run consecutively for up to 20 rounds until the task is truly finished
The author also mentions that /goal draft can organize messy requirements into something more like a "completion contract."
More from coding & agent
- A roundup of AI agents and MCP resources, including how to evaluate agents — _jaydeepkarale · 2026-07-21
- A full course shows how to build and deploy an AI agent with OpenAI and LangChain — _jaydeepkarale · 2026-07-21
- A beginner guide to AI agents points readers to a Stanford webinar — _jaydeepkarale · 2026-07-21
- A practical guide on how to evaluate AI agents — _jaydeepkarale · 2026-07-21
- MCP is headed toward easier scale, event-driven extensions, and workable file uploads — EricBuess · 2026-07-21
- Developers debate the missing composition model for AI agents — threepointone · 2026-07-21