agentleFS
jadmadijadmadi/dojo/skills/dojo/SKILL.md

dojo

Skill0 starsChanged 7 days agoApache-2.0View on GitHub

"Enter the Dojo!": Agent Skills Training Ground, Native MCP Gateway, and Multi-Harness Continuity Engine. Activate this skill to discover skills with hybrid BM25+RRF search, inspect Markdown section TOCs (`dojo toc`), JIT load specific #sections or full skills with dependency graphs (`dojo load`), bi-directionally synchronize skills across Antigravity/Claude/Cursor/Devin/OpenCode/Sila (`dojo sync` / `dojo sila`), link a Personal Skills Registry and sync portable `skills.json` profiles and local commands/skills between local machines and `agy-vm` (`dojo profile link|sync|pull|push|apply`), initialize workspaces (`dojo init --workspace`), scaffold and audit skills (`dojo init` / `dojo doctor`), run companion scripts (`dojo run`), or expose Dojo as a native MCP server (`dojo mcp`).

8 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions8 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 8

    1. Necmttn/ax★ 116same name
    2. a-ratz/dojo★ 3same name
    3. AndreRatzenberger/dojo★ 3same name
    4. guerrilla2799/sdr-copilot★ 1same name
    5. Iamkach/claude-dojo★ 1same name
    6. Lozadaa/sensei-dojo★ 0same name
    7. quangtran88/dojo★ 0same name
    8. supa-guru/skills★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

The file

Files
SKILL.md

Dojo — Agent Skills Training Ground & Multi-Harness Continuity Engine

dojo is a static Go CLI and native MCP server for finding, streaming, auditing, and bi-directionally synchronizing Agent Skills (SKILL.md) and agentic commands across local workstations and cloud VMs (agy-vm).

1. Search & Section-Level Paging (dojo find, dojo toc, dojo load)

Inspect a skill's Table of Contents first, then stream only the relevant #section so the agent reads what the task needs. dojo toc reports the byte size of every section before anything is loaded. A bundled benchmark enforces a 45 percent ceiling on the largest leaf section as a share of the body:

# Search catalogs with hybrid BM25 + fuzzy matching
dojo find "go testing"

# Inspect headings, #slugs, line ranges, and byte sizes
dojo toc godoctor

# Load only a single section by #slug or --section
dojo load godoctor#quick-start

# Load a skill with all transitive dependencies in topological order
dojo load fullstack-go --with-deps

2. Bi-Directional Multi-Harness Continuity (dojo sync / dojo sila)

dojo reconciles newly created or edited skills from any harness back into .agents/skills/<name>/SKILL.md (including reverse-transpiling Cursor .mdc rules) and projects canonical skills across all active harnesses:

  • Antigravity (Desktop & CLI): .agents/skills/<name> in a workspace, ~/.gemini/config/skills/<name> globally
  • Claude Code: .claude/skills/<name>
  • Cursor: .cursor/skills/<name> (.cursor/rules/<name>.mdc remains available with --target cursor-rules)
  • Devin (formerly Windsurf): .devin/skills/<name> (windsurf is an alias)
  • OpenCode: .opencode/skills/<name>
  • Sila: .sila/skills/<name>
# Learn a skill and immediately sync to all harnesses
dojo learn godoctor --with-deps --target all

# Bi-directionally reconcile and project skills across harnesses
dojo sila --target antigravity,cursor,claude

3. Personal Skills Registry & Local-to-Cloud (agy-vm) Continuity

Link a GitHub repository as your Personal Skills Registry once, then use dojo profile sync or dojo profile pull across local workstations and cloud VMs (agy-vm):

# 1. Link your personal skills registry (once per machine):
dojo profile link github:jadmadi/agy-vm -g

# 2. Two-way sync (reconciles local harnesses, pulls remote skills, updates skills.json, and pushes local skills to GitHub):
dojo profile sync

# 3. On agy-vm (or any target machine): pull and project all skills from your linked registry:
dojo profile pull -g

# 4. Push specific skills, list indices, rules (.mdc), or slash commands (.claude/commands/*.md):
dojo profile push <x> <y>
dojo profile push github:jadmadi/agy-vm ~/.claude/commands/x.md ~/.agents/skills/y

What dojo profile sync and dojo profile push search and package automatically:

  • Skill directories (including SKILL.md, scripts/, references/, assets/): .agents/skills/<name>, .gemini/skills/<name>, .antigravity/skills/<name>, .claude/skills/<name>, .opencode/skills/<name>, .sila/skills/<name> (in current workspace and ~/).
  • Rules & Slash Commands (auto-converted into compliant SKILL.md packages with frontmatter): .cursor/rules/<name>.mdc, .windsurf/rules/<name>.md, .claude/commands/<name>.md, .gemini/commands/<name>.md, .agents/workflows/<name>.md, .windsurf/workflows/<name>.md.
  • For complex multi-file command migrations, load the full reference guide: dojo load dojo/references/migration-guide.md.

4. Native MCP Server & Shell Completions (dojo mcp & dojo completion)

# Start JSON-RPC 2.0 MCP server on stdio or inspect tool schema
dojo mcp
dojo mcp --print-schema
# Tool results include JSON text and structuredContent. Read-only tools carry
# readOnlyHint; run, learn, sync, and profile carry destructive or open-world
# hints. learn, sync, and profile push accept --dry-run for a no-write preview.

# Enable dynamic shell tab-completion (skills, #sections, flags, targets)
source <(dojo completion bash)
source <(dojo completion zsh)

5. Skill Authoring, Linting & Security Auditing (dojo init & dojo doctor)

# Initialize a multi-harness workspace (.agents/skills, skills.json, and harness projections)
dojo init --workspace

# Scaffold a new compliant skill package (SKILL.md, scripts/, references/, evals/evals.json)
dojo init my-new-skill --category coding

# Audit config, frontmatter compliance, token budgets, and scan for prompt injection / unsafe shell pipes
dojo doctor --skills --security

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.