Skip to content
Development
Skill

/hermes-workspace

Use when laying out or cleaning the file workspace shared by a Hermes Agent chief and its managers - inbox, dated dossiers (README / received pieces / internal work / deliverables), clients, admin areas, knowledge, automation, archives and a to-delete area awaiting approval, a

From plugin
claude-plugins
36200 skills42 agents34 commands

Claude Code

$ npx -y skills add fusengine/claude-plugins --skill hermes-workspace --agent claude-code

SKILL.md

hermes-workspace.SKILL.md
name: hermes-workspace
description: Use when laying out or cleaning the file workspace shared by a Hermes Agent chief and its managers - inbox, dated dossiers (README / received pieces / internal work / deliverables), clients, admin areas, knowledge, automation, archives and a to-delete area awaiting approval, a mirrored private space, scripts kept apart from documents, scratch outside the workspace, file naming, the end-of-task checklist and a weekly cleanup job that only asks before deleting. Not for context-file resolution rules (hermes-core) or sandboxing (hermes-security).
versions:
  hermes-agent: "0.23.0"
user-invocable: false
references:
  - references/layout.md
  - references/naming-scripts-scratch.md
  - references/checklists.md
  - references/templates/hermes-md-filing.md
  - references/templates/dossier-readme.md
  - references/templates/cleanup-job.md
  - references/templates/knowledge-schema.md
related-skills:
  - hermes-core
  - hermes-cron
  - hermes-chief-of-staff
  - hermes-security
  - hermes-rhythm

<objective> One filing logic for the chief and every agent, written once in the shared `.hermes.md`: where each kind of document goes, how files are named, where scripts and temporary files live (never among documents), what every agent ticks before closing a task, and a weekly cleanup that tidies what is safe and asks before deleting anything. </objective>

Hermes Workspace

Freshness check

Before writing or changing any Hermes config, run `hermes --version` and `hermes update --check`, then read https://github.com/NousResearch/hermes-agent/releases/latest. If the installed or latest version is newer than 0.23.0, read every release note since 0.23.0 ("Upgrade notes" first) and adjust. Never apply guidance from this skill to a newer version without that check.

Hermes facts this skill relies on

| Fact | Consequence | Owner skill | |------|-------------|-------------| | `.hermes.md` is found by walking from the working directory up to the git root | Put the filing rules in `.hermes.md` at each workspace root; make that root the git root (or the cwd) of every agent | hermes-core | | Cron jobs load project context only with `--workdir` | Every workspace job (cleanup, rituals) passes `--workdir <root>` | hermes-cron | | `terminal.cwd` per profile | Set each profile's cwd inside its own root | hermes-core | | Kanban `scratch` workspaces are deleted on task completion; `dir:<abs path>` is preserved | Use `dir:` for dossier work, keep temporary files in scratch | hermes-delegation | | No native "write only here" policy | Containers / `HERMES_WRITE_SAFE_ROOT` + checklist | hermes-security |

Decision map

| Question | Load | |----------|------| | Which folder for what, synced chief root, nested `.hermes.md`, company vs private, transition symlinks | [layout.md](references/layout.md) | | Knowledge base (`SCHEMA.md`, `index.md`, `log.md`) | [templates/knowledge-schema.md](references/templates/knowledge-schema.md) | | File names, scripts, temporary files | [naming-scripts-scratch.md](references/naming-scripts-scratch.md) | | End-of-task checklist, weekly cleanup, who keeps order | [checklists.md](references/checklists.md) | | Filing block for `.hermes.md` (+ company outbound email block) | [templates/hermes-md-filing.md](references/templates/hermes-md-filing.md) | | Dossier README | [templates/dossier-readme.md](references/templates/dossier-readme.md) | | Weekly cleanup cron job | [templates/cleanup-job.md](references/templates/cleanup-job.md) |

Critical rules

1. **Documents and scripts never mix.** The workspace holds documents; each script has its own folder with a README, outside the document roots. 2. **Scratch lives outside the workspace** and is disposable. 3. **Received pieces are never modified**; work happens in `travail/` (internal notes) and `livrables/` (deliverables). 4. **Nothing is deleted by an agent.** Candidates go to the to-delete area; the owner approves deletion. 5. **Doubt about private vs company → ask the chief**, never create a new top-level folder. 6. **Same rules for every agent**: full reference in the chief root's `.hermes.md`, the applicable part repeated in each root's `.hermes.md`. 7. **A synced workspace is the owner's view**: no secrets, no scratch, one final version per deliverable. 8. **Old paths stay as transition symlinks**; nothing new is created under them.

Sources

  • https://hermes-agent.nousresearch.com/docs/user-guide/features/context-files
  • https://hermes-agent.nousresearch.com/docs/user-guide/features/cron (`--workdir`)
  • https://hermes-agent.nousresearch.com/docs/user-guide/features/kanban (workspace kinds)
  • https://hermes-agent.nousresearch.com/docs/user-guide/configuration (`terminal.cwd`)
  • https://hermes-agent.nousresearch.com/docs/user-guide/security (file write safety)
Read more
@fusengine
Ships withclaude-plugins

A plugin ecosystem that turns Claude Code into a supervised, multi-agent development environment.

36200 skills42 agents
Get the whole plugin

Stars on GitHub

Momentum

stars in total
36
Star history of fusengine/agents on GitHub, a chart by star-history.com

Other skills on claude-plugins.

brainstorming

brainstorming

Use when creating a feature/component or adding functionality. Fires BEFORE APEX Analyze to refine requirements via structured…

challenge

challenge

Use before a root-cause, done/verified claim, irreversible action, or 2nd-time fix reaches the owner (APEX or plain…

code-quality

code-quality

Use when validating code quality after modifications -- SOLID compliance, DRY duplication, linter errors, architecture…

elicitation

elicitation

Use when an expert agent self-reviews and self-corrects code after the Execute phase, before sniper validation (BMAD-METHOD…

More skills from other plugins.

brainstorming
Auto-invoked

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or…

@obra@obraInstall Skill
ask-matt

ask-matt

Ask which skill or flow fits your situation. A router over the skills in this repo.

skills2w309k
@mattpocock@mattpocockInstall Skill
accessibility
Auto-invoked

accessibility

Design, implement, and audit accessible UI to WCAG 2.2 Level AA across Web, iOS, and Android — semantic ARIA roles and labels,…

ecc2w293k
@affaan-m@affaan-mInstall Skill
compiler-commit

compiler-commit

Use when you want to verify compiler changes and commit with the correct convention. Runs tests, lint, and format, then commits…

react5d252k
@facebook@facebookInstall Skill

Questions about hermes-workspace.

Which coding agents can use hermes-workspace?
hermes-workspace installs in Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, OpenCode and Windsurf with the skills CLI, using each agent's flag shown on this page. Flowy's auto-invocation, which fires a skill when your prompt matches, runs in Claude Code only.