docs: add plugin systems comparison (ours vs claude vs emes)
Compares three approaches: - Our system: cross-agent, Nix, lenses - Claude plugins: official, hooks, marketplace - emes: mechanical enforcement, tissue, idle, jwz Living document for iterating on architecture. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This commit is contained in:
dan
parent
a84066c1dd
commit
8a76f4e9cb
135
docs/research/plugin-systems-comparison.md
Normal file
135
docs/research/plugin-systems-comparison.md
Normal file
|
|
@ -0,0 +1,135 @@
|
|||
# Plugin Systems Comparison: Ours vs Claude vs emes
|
||||
|
||||
> **Date:** 2026-01-09
|
||||
> **Status:** Living document
|
||||
> **Related:** [ADR-005](../adr/005-dual-publish-plugin-architecture.md)
|
||||
|
||||
## Overview
|
||||
|
||||
Three plugin/skill systems for AI coding agents:
|
||||
|
||||
| System | Origin | Focus |
|
||||
|--------|--------|-------|
|
||||
| **Ours** | This repo | Cross-agent portability |
|
||||
| **Claude Plugins** | Anthropic (Oct 2025) | Claude Code extensibility |
|
||||
| **emes** | [evil-mind-evil-sword](https://github.com/evil-mind-evil-sword) | Mechanical quality enforcement |
|
||||
|
||||
## Feature Comparison
|
||||
|
||||
| Aspect | **Our System** | **Claude Plugins** | **emes** |
|
||||
|--------|---------------|-------------------|----------|
|
||||
| Primary file | `SKILL.md` | `plugin.json` + `skills/*.md` | `plugin.json` + tools |
|
||||
| Distribution | Nix + `deploy-skill.sh` | `/plugin install` + marketplace | `/plugin install @emes` |
|
||||
| Discovery | Symlink to `~/.claude/skills/` | Auto-discovery from `skills/` | Auto-discovery |
|
||||
| Multi-agent | ✅ Gemini, OpenCode, Claude | ❌ Claude only | ❌ Claude only |
|
||||
| Hooks | ❌ None | ✅ 9 lifecycle events | ✅ Uses hooks heavily |
|
||||
| Issue tracking | beads | N/A | tissue (their own) |
|
||||
| Quality gates | ❌ Manual | ✅ Hooks possible | ✅ idle (blocks exit) |
|
||||
| Messaging | N/A | N/A | jwz (async, identity-aware) |
|
||||
|
||||
## Architectural Principles
|
||||
|
||||
| Principle | Ours | Claude | emes |
|
||||
|-----------|------|--------|------|
|
||||
| Context injection | Push (SKILL.md loaded upfront) | Push (skill loaded upfront) | **Pull on-demand** |
|
||||
| Enforcement | Prompt-based ("please do X") | Hooks available but optional | **Mechanical (hooks required)** |
|
||||
| Content style | Inline markdown | Inline markdown | **References to files** |
|
||||
| Interface | Human-readable | Human-readable | **Machine-first (JSON)** |
|
||||
|
||||
## Unique Strengths
|
||||
|
||||
### What We Have (that others don't)
|
||||
|
||||
1. **Cross-agent support** - Works with Gemini, OpenCode, not just Claude
|
||||
2. **Nix deployment** - Declarative, reproducible, system-level
|
||||
3. **Lenses pattern** - Multi-pass focused review (bloat, security, coupling, etc.)
|
||||
4. **beads integration** - Issue tracking across sessions
|
||||
|
||||
### What Claude Has (that others don't)
|
||||
|
||||
1. **Official support** - Anthropic-maintained, documented
|
||||
2. **Marketplace ecosystem** - Community plugins, discovery
|
||||
3. **Scope control** - user/project/local installation
|
||||
4. **9 lifecycle hooks** - SessionStart, PostToolUse, PreCompact, etc.
|
||||
|
||||
### What emes Has (that others don't)
|
||||
|
||||
1. **idle** - Quality gate that mechanically blocks exit until reviewer approves
|
||||
2. **tissue** - Issue tracker designed for agents (JSON, non-interactive)
|
||||
3. **jwz** - Async messaging with git context and identity
|
||||
4. **alice** - Adversarial reviewer agent (Opus-based)
|
||||
5. **Mechanical enforcement** - Hooks enforce policy, not prompts
|
||||
|
||||
## emes Tool Details
|
||||
|
||||
| Tool | Purpose | Language |
|
||||
|------|---------|----------|
|
||||
| **idle** | Quality gate - blocks exit until alice reviews | Shell + hooks |
|
||||
| **tissue** | Git-native issue tracking, machine-first | Zig |
|
||||
| **jwz** | Async messaging with identity/git context | Zig |
|
||||
| **marketplace** | Plugin distribution registry | JSON |
|
||||
| **alice** | Read-only Opus reviewer, creates issues | Agent |
|
||||
|
||||
### emes Key Principles
|
||||
|
||||
From their architecture:
|
||||
|
||||
1. **Pull context on-demand** - Don't inject large prompts upfront
|
||||
2. **Mechanical enforcement** - Use hooks to enforce, not prompt instructions
|
||||
3. **References over inline** - Point to files, don't embed content
|
||||
4. **Machine-first interfaces** - JSON output, non-interactive CLIs
|
||||
|
||||
## Philosophical Stances
|
||||
|
||||
| System | Core Belief |
|
||||
|--------|-------------|
|
||||
| **Ours** | "Skills should work everywhere" - agent-agnostic |
|
||||
| **Claude** | "Plugins extend Claude Code" - platform-specific |
|
||||
| **emes** | "Enforce quality mechanically" - trust hooks, not prompts |
|
||||
|
||||
## Integration Points
|
||||
|
||||
### We Could Adopt from emes
|
||||
|
||||
- [ ] **idle pattern** - Quality gate for code-review skill (skills-9jk)
|
||||
- [ ] **tissue** - Alternative to beads? (probably not, beads is good)
|
||||
- [ ] **Pull on-demand** - Reduce upfront context injection
|
||||
- [ ] **Machine-first** - JSON output modes for scripts
|
||||
|
||||
### We Could Adopt from Claude
|
||||
|
||||
- [x] **plugin.json** - Already doing via dual-publish
|
||||
- [x] **marketplace.json** - Already created
|
||||
- [ ] **Hooks** - Add to skills that need lifecycle events
|
||||
- [ ] **Scope control** - Let users choose install scope
|
||||
|
||||
## Current Strategy: Dual-Publish
|
||||
|
||||
We maintain both systems:
|
||||
|
||||
```
|
||||
skills/my-skill/
|
||||
├── .claude-plugin/
|
||||
│ └── plugin.json # Claude plugin system
|
||||
├── skills/
|
||||
│ └── my-skill.md # Claude auto-discovery
|
||||
├── SKILL.md # Nix deployment (cross-agent)
|
||||
└── README.md
|
||||
```
|
||||
|
||||
**Rationale:** Cross-agent support is important. Gemini can't use Claude plugins (skills-bo8).
|
||||
|
||||
## Open Questions
|
||||
|
||||
1. Should we adopt tissue or stick with beads?
|
||||
2. How much of emes philosophy (mechanical enforcement) should we adopt?
|
||||
3. Can we make lenses work as Claude plugin hooks?
|
||||
4. Should skills output JSON for machine consumption?
|
||||
|
||||
## References
|
||||
|
||||
- [Claude Code Plugin Docs](https://code.claude.com/docs/en/plugins)
|
||||
- [emes org](https://github.com/evil-mind-evil-sword)
|
||||
- [emes idle](https://github.com/evil-mind-evil-sword/idle)
|
||||
- [emes tissue](https://github.com/evil-mind-evil-sword/tissue)
|
||||
- [Anthropic marketplace](https://github.com/anthropics/claude-code/blob/main/.claude-plugin/marketplace.json)
|
||||
Loading…
Reference in a new issue