agentleFS
PuntoyComaTechPuntoyComaTech/hermes-agent-squads/skills/managing-projects/SKILL.md

managing-projects

Skill3 starsChanged 18 days agoMITView on GitHub

Creates squad Projects and copies them to the workers. Use when a squad starts working in a new project folder.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    When the folder ships .hermes/skills/ or .agents/skills/, show the user those skill names and, after an explicit OK, run hermes -p <bot> …

    line 31

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Managing projects

When to Use

  • A squad takes on a new site, brand, client or search: it becomes a Project.
  • The user asks where a project's sessions or worktrees live.

Procedure

  1. Folder: the user picks it, new or an existing repo. Never choose one for them, never the home root or a profile directory.
  2. The user creates it: Desktop Sessions, the manager in the bottom selector, "+", then chats inside it (the chat's working directory is the project's primary folder). When the user names a folder in chat instead, a manager chat tool_calls desktop_project with {"action": "create", "name": "<name>", "path": "<their folder>"}. Otherwise: hermes -p <manager> project create "<name>" <folder>. Gateway channels (Discord, etc.) do not inherit the Desktop project: the channel prompt must name the folder.
  3. Copy it to the workers: scripts/hpy project_sync.py --manager <manager> (a manager runs python3 $HERMES_HOME/scripts/project_sync.py; add --dry-run to preview). It creates the project in every worker that lacks it and binds the squad board. A project made with "+" syncs the same way.
  4. Tell the user once: "Sessions: pick <manager> in the bottom selector to see its projects; + creates another in a folder you choose."
  5. When the folder ships .hermes/skills/ or .agents/skills/, show the user those skill names and, after an explicit OK, run hermes -p <bot> skills trust <path> for the manager and every worker. Untrusted project skills never load.
  6. Cards then take --project <slug>.
  7. Later changes (add-folder, set-primary, rename, archive): run hermes -p <bot> project <action> ... for the manager and every worker.

Pitfalls

  • Creating the project in the manager only: Project state is per profile, so workers report "no projects yet" until project_sync.py runs.
  • Writing project rows into a database by hand: use hermes project.
  • Choosing a folder for the user: a bot-chosen folder lands outside the project the user expects and leaves stray folders.
  • Expecting project_sync.py to create folders or set repo_scan_roots: --setup (run by squadctl apply) only resets the repo_scan_roots: [~/Projects/<squad>] that squad versions before 2.5.0 set.
  • Assuming a repo/ subfolder: a folder that already is the git repo keeps its layout; read the repo path from the squad's config (web: site.yaml -> repo.path).
  • Sweeping a shared parent folder with a sync, cleanup or scan: target the exact project paths. A parent that also holds the user's own repos treats them as squad data.

Verification

  • hermes -p <bot> project list shows the project in the manager and each worker.
  • A trusted repo: hermes -p <bot> config get skills.trusted_project_dirs lists its path in the manager and each worker.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.