Hermes Plan Mode — Design skill for Claude Code
Claude Code-style plan mode for Hermes Agent: enforced read-only planning, then approval through Hermes' own approval prompt on every interactive surface.
How to install Hermes Plan Mode
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open 100yenadmin/hermes-plan-mode and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Hermes Plan Mode does
Claude Code-style plan mode for Hermes Agent: enforced read-only planning, then approval through Hermes' own approval prompt on every interactive surface.
Alternatives in Design
- Learning Output Style — Interactive learning mode for skill building 14k ★
- Polish UI — Polish UI elements by auditing SCSS and TSX markup against the Anytype design system, then fixing discrepancie 8.7k ★
- Plannotator — Interactive plan review UI for Claude Code with visual annotation, offline sharing, and Obsidian/Bear integrat 3.3k ★
README
Hermes Plan Mode

Claude Code / Codex-style plan mode for Hermes, on every interactive Hermes surface (CLI, TUI, Desktop and the messaging platforms), with no core changes. The agent explores with read-only tools, asks clarifying questions, writes a plan file and shows it to you. You approve it with Hermes' own approval prompt: buttons, a card, the CLI panel or `/approve`. Then the agent implements it in the same turn and tracks progress. Until you approve, the plugin blocks mutating tools dispatched through Hermes' `pre_tool_call` hook, except writes to the plan file.
`/plan` in core Hermes is a prompt only. This plugin turns it into an enforced mode.
How it works
- Start planning with
/planor/planmode on [task]. The agent can also enter on its own with theplan_modetool. - The agent explores with read-only tools. For a choice only you can make, it asks with
clarifybefore it writes the plan. With the defaultleanprofile it prefers one question and asks at most three, each with 2–3 choices, recommended first. - It writes the plan as Markdown under
/.hermes/plans. - It shows the complete plan in its reply, then calls
plan_mode(action="submit"). - Hermes shows its normal approval prompt for plan revision N: buttons, a card, the CLI panel, or
/approvetext. - Approve: plan mode turns off and the agent implements the plan in the same turn, mirroring the steps into
todo_list. - Deny: plan mode stays on. The plugin tells the agent that a denial is review feedback, so it asks what to change (when no reason came with the denial), revises the plan and submits a complete new revision.
/plan ─┐
/planmode on ─┼─► PLANNING ── read-only tools, clarify, write the plan file
agent: on ─┘ │
│ agent shows the plan, calls plan_mode(submit)
▼
Hermes approval prompt (rev N)
Related Skills
Product Ideation To Planning
Transform a product vision through Discover → Define → Design → Plan so the team is teed up for a fast Build w
Sdd Ff.Prompt
Fast-forward SDD planning through proposal, specs, design, and tasks.
Unity Mockup
Produces a visual mockup for a GitHub issue using the design canvas skill, matching the project's existing vis
Skuggsja
A local, read-only retrospective for Claude Code, Codex, Hermes and Cursor histories. One binary, embedded UI,
BotServer
Telegram/Discord/Slack bot that drives Claude through Anthropic API, Claude Code CLI, or Claude Desktop UI aut
Sesori Figma Review
Figma plugin + local Claude Agent SDK bridge: Claude reviews your prototype inside Figma, flow then screen by
Related Agents
Harness Runtime Architect
Reviews the harness library/CLI surface — plan/policy/approval/execute/audit/rollback semantics, evidence bund
Antigravity Agent
Runs Google Antigravity on the user's Antigravity subscription, reached through the agy CLI. One mode - consul
Agent Safety Reviewer
Read-only review of model-facing behaviour: tool risk classification, approval gating and its UX, prompt-injec