Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add rules/williamisnotdefined/rubiks-cube-solver/solver-searchgit clone --depth 1 https://github.com/williamisnotdefined/rubiks-cube-solverWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.02989 |
| Opus 5 | $0.00000 | $0.01494 |
| Sonnet 5 | $0.00000 | $0.00598 |
| Haiku 4.5 | $0.00000 | $0.00299 |
Grade A, and why
solver-search scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 292 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Generated from ai/registry.json. Do not edit manually.
Canonical skill: ../../ai/skills/solver-search.md.
Referenced context:
../../ai/rules/solver-search-rules.md../../ai/rules/testing-rules.md../../ai/architecture/project-architecture.md../../ai/architecture/solver-search.md../../ai/architecture/cube-engine.md../../ai/glossary/cube-terms.md
This file is compiled from canonical AI knowledge files. Edit canonical files under ai, then run npm run ai:sync.
Compiled AI Skill: solver-search
Canonical Skill: ai/skills/solver-search.md
Solver Search
Use this skill when adding or changing IDA*, generated two-phase search, heuristics, pruning, or pattern database code.
Goal
Add search behavior only on top of correct cube state and move semantics.
Read First
ai/rules/solver-search-rules.mdai/rules/testing-rules.mdai/architecture/project-architecture.mdai/architecture/solver-search.mdai/architecture/cube-engine.mdai/glossary/cube-terms.md
Workflow
- Verify cube move behavior before trusting search results.
- Start with shallow deterministic cases.
- Keep heuristics separate from traversal logic.
- Mark whether each heuristic is admissible.
- Add node/depth metrics for non-trivial algorithms.
Expected Output
- Search returns explicit move sequences and metrics.
- Tests cover solved state and shallow known scrambles.
- No learned or external inference heuristic is introduced without an explicit current product requirement.
Verification
- Run
cargo test -p cube-enginewhen Rust is installed. - Include targeted tests for the changed search module.
Referenced Context
Reference: ai/rules/solver-search-rules.md
Solver Search Rules
Rules for search, heuristics, and pattern database work.
Always
- Keep search algorithms in
crates/cube-engine/src/searchunless a later crate boundary is introduced. - Keep heuristics explicit about admissibility.
- Prefer bounded IDA* and generated two-phase paths after move tables are correct.
- Prune inverse moves and repeated same-axis branches where correctness allows it.
- Measure node counts and depth limits for non-trivial search changes.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 292 lines · 0 tokens per session scan A bf2434c70c44
solver-search is a cursor rule published in the GitHub repository williamisnotdefined/rubiks-cube-solver (4 stars, last pushed 9d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,989 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other cursor rules, from other repositories
08-client-server
IF 判断当前环境是否为客户端 → FabricLoader.getInstance().getEnvironmentType() == EnvType.CLIENT → World.isClient（Yarn 字段；yarn 1.14.4 World.mapping）。不要用 MCP/Forge 的 isRemote.
07-datagen
IF 生成物品/方块模型 JSON → 手动编写 JSON 在 src/main/resources/assets/{modid}/models/.
meta-quest-agentic-tools
Use Meta Quest Agentic Tools for Meta Quest and Horizon OS samples.
webgpu-threejs-tsl
Entry point for Three.js WebGPU + TSL guidance. Covers renderer setup, node materials, and TSL syntax fundamentals. Auto-attaches on JS/TS files in projects using three/webgpu.
serialization-rules
Unity serialization best practices.
unreal-networking
Networking and multiplayer in Unreal Engine.