[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-83633":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":15,"subscribersCount":15,"size":15,"stars1d":15,"stars7d":15,"stars30d":15,"stars90d":15,"forks30d":15,"starsTrendScore":15,"compositeScore":16,"rankGlobal":10,"rankLanguage":10,"license":17,"archived":18,"fork":18,"defaultBranch":19,"hasWiki":18,"hasPages":18,"topics":20,"createdAt":10,"pushedAt":10,"updatedAt":27,"readmeContent":28,"aiSummary":10,"trendingCount":15,"starSnapshotCount":15,"syncStatus":29,"lastSyncTime":30,"discoverSource":31},83633,"rp-engine","jieyefriic\u002Frp-engine","jieyefriic","YAML-native agent workflow execution engine, written in Rust","https:\u002F\u002Fcrates.io\u002Fcrates\u002Friceprompt-engine",null,"Rust",1238,8,20,0,15.86,"Apache License 2.0",false,"main",[21,22,23,24,25,26],"agent","llm","mcp","rust","workflow","yaml","2026-06-12 02:04:35","# riceprompt-engine\n\nYAML-native agent workflow execution engine, written in Rust.\n\nYou describe an agent workflow as a YAML file — nodes, edges, prompts, data\nsources, MCP tools — and the engine parses it, resolves dependencies, and\nexecutes the graph: making LLM calls, running scripts, querying databases,\ncalling MCP tools, iterating over data, and orchestrating multi-agent plans.\n\nThis engine powers **[RicePrompt](https:\u002F\u002Friceprompt.app)** — the visual\nagent IDE where you build and run these workflows without writing YAML by hand.\n\n```toml\n[dependencies]\nriceprompt-engine = \"0.1\"\n```\n\n## Features\n\n- **YAML-native** — entire workflow (graph, prompts, data sources, providers)\n  in a single declarative file. See [`docs\u002FFLOW_SPEC.md`](docs\u002FFLOW_SPEC.md)\n  for the authoritative spec.\n- **Multi-provider LLM support** — OpenAI, Anthropic, Gemini, DeepSeek, Qwen,\n  Zhipu, Moonshot, MiniMax, xAI, Huoshan, and any OpenAI-compatible endpoint.\n- **Streaming, tool calling, structured output** — first-class across providers.\n- **Rich node types** — `generate`, `transform` (Rhai scripting), `iterator`,\n  `supervisor` (multi-agent routing), `subgraph`, `data_connector`,\n  `skill_set` (progressive-disclosure knowledge bundles), `mcp` \u002F\n  `mcp_tools` (Model Context Protocol).\n- **Built-in data connectors** — PostgreSQL, MySQL, MongoDB, Redis, Qdrant,\n  S3-compatible object storage, REST APIs.\n- **Harness layer** — workflow-level instructions (CLAUDE.md-style) injected\n  into every generate node, with persistent memory support.\n- **Self-describing results** — `ExecutionResult` can include the source YAML\n  so downstream tooling renders the topology + per-node results from one file.\n- **Checkpoint \u002F resume** — pause and resume long-running workflows.\n\n## Quick start\n\nA minimal three-node workflow:\n\n```yaml\nversion: \"1.0\"\nname: \"hello_world\"\n\nproviders:\n  openai:\n    api_key: \"${OPENAI_API_KEY}\"\n\nnodes:\n  - id: start\n    type: start\n\n  - id: greet\n    type: generate\n    config:\n      provider: openai\n      model: gpt-4o-mini\n      template: tpl_greet\n      variables:\n        name: \"start.name\"\n\n  - id: response\n    type: response\n    config:\n      output:\n        greeting: \"greet.output\"\n\nedges:\n  - from: start\n    to: greet\n  - from: greet\n    to: response\n\ntemplates:\n  tpl_greet:\n    user_prompt: \"Greet {{name}} warmly in one sentence.\"\n```\n\nRun it:\n\n```rust\nuse riceprompt_engine::Engine;\nuse serde_json::json;\n\n#[tokio::main]\nasync fn main() -> anyhow::Result\u003C()> {\n    let yaml = std::fs::read_to_string(\"hello.yaml\")?;\n    let engine = Engine::builder().build()?;\n    let result = engine.run_yaml(&yaml, json!({ \"name\": \"Ada\" })).await?;\n    println!(\"{}\", serde_json::to_string_pretty(&result)?);\n    Ok(())\n}\n```\n\nMore runnable examples live under [`examples\u002F`](examples\u002F).\n\n## Documentation\n\n- [`docs\u002FFLOW_SPEC.md`](docs\u002FFLOW_SPEC.md) — authoritative YAML workflow spec\n  (node types, fields, providers, data sources, harness, skills, MCP).\n- A user-facing usage guide (\"skill guide\") will be published separately.\n\n## Related\n\n- **[RicePrompt](https:\u002F\u002Friceprompt.app)** — visual agent IDE built on top\n  of this engine. Design workflows in a graph editor, run them in-browser,\n  and export the same YAML this engine consumes.\n\n## Project status\n\n`0.1.x` — the API may change between minor versions while the spec\nstabilizes. Pin an exact version if you need stability.\n\n## Contributing\n\nIssues and PRs welcome. Please:\n\n- Run `cargo fmt` and `cargo clippy --all-targets` before submitting.\n- Add tests for new node types or provider behaviors.\n- For changes that touch the YAML surface, update `docs\u002FFLOW_SPEC.md` in\n  the same PR.\n\n## License\n\nLicensed under either of\n\n- Apache License, Version 2.0 ([LICENSE-APACHE](LICENSE-APACHE) or\n  \u003Chttp:\u002F\u002Fwww.apache.org\u002Flicenses\u002FLICENSE-2.0>)\n- MIT license ([LICENSE-MIT](LICENSE-MIT) or\n  \u003Chttp:\u002F\u002Fopensource.org\u002Flicenses\u002FMIT>)\n\nat your option.\n\nUnless you explicitly state otherwise, any contribution intentionally\nsubmitted for inclusion in the work by you, as defined in the Apache-2.0\nlicense, shall be dual licensed as above, without any additional terms or\nconditions.\n",2,"2026-06-11 04:11:24","high_star"]