📄 skillActivePromisingmedium
statewright/statewright
by statewright
State machine guardrails for AI agents
Supported Platforms
🤖 Claude Code
Quality Breakdown
92/ 200
Content Signals
Repo Health
Multi-platform bonus: +5 pts if tool supports 2+ platforms. Score derived from 12 structural signals — not stars or popularity.
Trust & Verification
medium
Requires extended permissions (shell access, subagents). Review before use.
Active
Updated within the last 90 days. Actively maintained.
Risk Assessment
- MCP Gateway component includes bash classification and command enforcement logic (bash_classifier.rs, enforcement.rs) that intercepts and controls shell execution
- Repository contains orchestrator and executor modules that spawn and manage agent execution flows (orchestrator.rs, executor.rs)
- Multi-agent coordination capabilities present (harmony.rs, agent crate design suggests multi-model orchestration)
- Plugin marketplace integration (.agents/plugins/marketplace.json, .claude-plugin/) indicates external tool loading mechanism
- Persistent configuration storage in .claude/ directory (CLAUDE.md) for agent state and workflow management
- MCP Gateway acts as a proxy/interceptor for tool calls, which could be exploited if gateway logic has bypasses