agentleFS
Sign inStart free
cukascukas/claudes-ai-buddies/skills/add-buddy/SKILL.md

add-buddy

Skill9 starsChanged 8 months agoMITView on GitHub

Register a new AI CLI as a buddy — interactive wizard or one-liner

  • Installs packages

The file

We split this skill into the rules it gives an agent

12 rules. About 0 are common, 11 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈11

    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. Distinctive:

    Run the registration script directly:

    line 26
  2. Distinctive:

    Verify the buddy was registered:

    line 38
  3. Distinctive:

    Check if the binary is available:

    line 44
  4. Distinctive:

    Ask for the binary name (what to type on the command line to run it).

    line 55
  5. Distinctive:

    Ask for display name (defaults to the ID).

    line 56
  6. Very specific:

    Ask which modes it supports: exec (run tasks), review (code review), or both.

    line 57
  7. Distinctive:

    Ask for install hint (optional — shown when the binary isn't found).

    line 58
  8. Distinctive:

    Ask for timeout (default 360s).

    line 59
  9. Distinctive:

    Run the registration script with gathered info.

    line 60
  10. Distinctive:

    Test if the binary exists on PATH.

    line 61
  11. Distinctive:

    If the binary isn't found:

    line 76
  12. Distinctive:

    Test the binary: after registration. Report found/not-found status.

    line 86

≈ “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

/add-buddy — Register a New AI Buddy

Register any CLI-based AI tool as a buddy so it participates in /forge, /brainstorm, and /tribunal.

How to invoke

One-liner:

/add-buddy --id aider --binary aider --display "Aider" --modes exec

Interactive (just the name):

/add-buddy aider

Step-by-step workflow

If args are complete (--id and --binary provided):

  1. Run the registration script directly:
bash "${CLAUDE_PLUGIN_ROOT}/scripts/buddy-register.sh" \
  --id "ID" \
  --binary "BINARY" \
  --display "DISPLAY_NAME" \
  --modes "MODES" \
  --install-hint "HINT" \
  --timeout SECS
  1. Verify the buddy was registered:
cat "${HOME}/.claudes-ai-buddies/buddies/ID.json"
  1. Check if the binary is available:
source "${CLAUDE_PLUGIN_ROOT}/hooks/lib.sh"
ai_buddies_find_buddy "ID" && echo "Found" || echo "Not found"
  1. Report the result.

If interactive (only name/id provided):

  1. Ask for the binary name (what to type on the command line to run it).
  2. Ask for display name (defaults to the ID).
  3. Ask which modes it supports: exec (run tasks), review (code review), or both.
  4. Ask for install hint (optional — shown when the binary isn't found).
  5. Ask for timeout (default 360s).
  6. Run the registration script with gathered info.
  7. Test if the binary exists on PATH.
  8. Report success.

Output format

Registered buddy 'aider':
  Binary: aider
  Display: Aider
  Modes: exec, review
  Status: FOUND on PATH (/usr/local/bin/aider)

The buddy will now appear in /forge, /brainstorm, and /tribunal.

If the binary isn't found:

  Status: NOT FOUND — install with: pip install aider-chat
  The buddy is registered but won't participate until the binary is available.

Rules

  • ID must be alphanumeric (plus hyphens/underscores). Reject anything else.
  • Don't overwrite builtin buddies (claude, codex, gemini, opencode). Warn and ask for a different ID.
  • Test the binary after registration. Report found/not-found status.
  • Never pass secrets in registration.

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.

Check your setup, free