try-works/role-model / .agents/skills/recursive-mode
recursive-mode
UnverifiedRepository workflow orchestration skill for staged implementation, locked artifacts, late-phase receipts, and durable memory maintenance. Use when executing recursive-mode runs, resuming a run, locking a phase, or verifying locks.
references/benchmarks/README.md
View on GitHub# Packaged benchmark scenarios
These fixtures are the packaged benchmark projects used by `scripts/run-recursive-benchmark.py`.
| Scenario | Tier | Purpose |
| --- | --- | --- |
| `local-first-planner` | easy | Local-first work-item planner with CRUD, filters, summary metrics, import/export, and screenshot-friendly UI validation. |
| `team-capacity-board` | medium | Team planning board with owners, capacity points, dependency awareness, risk/blocked states, and broader planning summaries. |
| `release-readiness-dashboard` | hard | Local release-control dashboard with milestones, blockers, incidents, readiness scoring, and richer derived views. |
| `scientific-calculator-rust` | xhard | Browser-hosted Rust/WASM scientific calculator that starts from a bootstrap-only dependency scaffold so the agent must author the real expression engine, UI, and browser behavior from near-zero product code. |
All packaged scenarios stay browser-local, runnable from disposable folders, and free of external databases or services.
Current packaged stacks:
- React + TypeScript + Vite for the easy/medium/hard fixtures
- Rust + WebAssembly with Trunk for the xhard fixture
Benchmark reports combine:
- heuristic rubric coverage
- the mandatory controller-side judge metric
The default combined benchmark score weights heuristic coverage at 70% and judge review at 30%, while still reporting raw build/test/preview outcomes, recursive-on worktree isolation status, and the underlying component scores separately.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 470k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |