How to install flyto-core?
flytohub/flyto-core
$git clone https://github.com/flytohub/flyto-coreInstalls into the current project.
›Prefer a prompt? Paste this to your agent
Install flyto-core by running `git clone https://github.com/flytohub/flyto-core`, then use it for the current task and follow its documentation at https://github.com/flytohub/flyto-core.
README.md
| 1 | [](https://mseep.ai/app/flytohub-flyto-core) |
| 2 | [](https://mseep.ai/app/9a708224-9666-46b6-8660-dad08fb16096) |
| 3 | |
| 4 | # Flyto2 Core - Open-Source AI Agent Framework and Workflow Automation Engine |
| 5 | |
| 6 | [](https://pypi.org/project/flyto-core/) |
| 7 | [](https://opensource.org/licenses/Apache-2.0) |
| 8 | [](https://www.python.org/downloads/) |
| 9 | |
| 10 | <!-- mcp-name: io.github.flytohub/flyto-core --> |
| 11 | |
| 12 | > **The open-source execution engine for AI agents. 451 modules, MCP-native, triggers, queue, versioning, metering.** |
| 13 | > |
| 14 | > **[flyto2.com](https://flyto2.com)** · [Cloud Automation](https://flyto2.com/cloud/) · [Documentation](https://docs.flyto2.com) · [MCP Docs](https://docs.flyto2.com/mcp/) · [YouTube](https://www.youtube.com/@Flyto2) |
| 15 | |
| 16 | Flyto2 Core is the open-source runtime behind Flyto2. It is built for people who |
| 17 | want an **AI agent framework** that actually runs work: browser automation, API |
| 18 | integration, web scraping, MCP server automation, replayable YAML recipes, |
| 19 | evidence capture, and deterministic tools that agents can call without |
| 20 | inventing unreviewed code. |
| 21 | |
| 22 | Use it when the question is simple but the job is annoying: "open this page, |
| 23 | capture the proof, extract the data, check performance, and let me retry only |
| 24 | the failed step." Flyto2 Core gives you a local execution engine for browser |
| 25 | automation, workflow replay, AI-agent tool calls, Web Vitals checks, screenshot |
| 26 | capture, structured extraction, and audit-ready evidence. |
| 27 | |
| 28 | The current public inventory is **451 registry-backed modules** across **84 |
| 29 | catalog categories**, including triggers, queue modules, workflow versioning, |
| 30 | metering hooks, browser automation, API calls, data transforms, verification, |
| 31 | files, and crypto. |
| 32 | |
| 33 | Good fit if you searched for: |
| 34 | |
| 35 | - open source AI agent framework for production workflows |
| 36 | - Python AI workflow automation with Playwright |
| 37 | - MCP server automation with trace and replay |
| 38 | - browser automation that can resume from a failed step |
| 39 | |
| 40 | ### Try in 30 seconds |
| 41 | |
| 42 | ```bash |
| 43 | pip install flyto-core[browser] && playwright install chromium |
| 44 | flyto recipe competitor-intel --url https://github.com/pricing |
| 45 | ``` |
| 46 | |
| 47 | ``` |
| 48 | Step 1/12 browser.launch ✓ 420ms |
| 49 | Step 2/12 browser.goto ✓ 1,203ms |
| 50 | Step 3/12 browser.evaluate ✓ 89ms |
| 51 | Step 4/12 browser.screenshot ✓ 1,847ms → saved intel-desktop.png |
| 52 | Step 5/12 browser.viewport ✓ 12ms → 390×844 |
| 53 | Step 6/12 browser.screenshot ✓ 1,621ms → saved intel-mobile.png |
| 54 | Step 7/12 browser.viewport ✓ 8ms → 1280×720 |
| 55 | Step 8/12 browser.performance ✓ 5,012ms → Web Vitals captured |
| 56 | Step 9/12 browser.evaluate ✓ 45ms |
| 57 | Step 10/12 browser.evaluate ✓ 11ms |
| 58 | Step 11/12 file.write ✓ 3ms → saved intel-report.json |
| 59 | Step 12/12 browser.close ✓ 67ms |
| 60 | |
| 61 | ✓ Done in 10.3s — 12/12 steps passed |
| 62 | ``` |
| 63 | |
| 64 | Screenshots captured. Performance metrics extracted. JSON report saved. **Every step traced.** |
| 65 | |
| 66 | <p align="center"> |
| 67 | <img src="demo/flyto-core-demo.gif" alt="flyto-core demo: API pipeline → replay → browser automation" width="720"> |
| 68 | </p> |
| 69 | |
| 70 | --- |
| 71 | |
| 72 | ## What happens when step 8 fails? |
| 73 | |
| 74 | With a shell script you re-run the whole thing. With flyto-core: |
| 75 | |
| 76 | ```bash |
| 77 | flyto replay --from-step 8 |
| 78 | ``` |
| 79 | |
| 80 | Steps 1–7 are instant. Only step 8 re-executes. Full context preserved. |
| 81 | |
| 82 | --- |
| 83 | |
| 84 | ## 3 recipes to try now |
| 85 | |
| 86 | ```bash |
| 87 | # Competitive pricing: screenshots + Web Vitals + JSON report |
| 88 | flyto recipe competitor-intel --url https://competitor.com/pricing |
| 89 | |
| 90 | # Full site audit: SEO + accessibility + performance |
| 91 | flyto recipe full-audit --url https://your-site.com |
| 92 | |
| 93 | # Web scraping → CSV export |
| 94 | flyto recipe scrape-to-csv --url https://news.ycombinator.com --selecto |
Related
6 picksType
- langchain-mcp-adaptersMake Anthropic Model Context Protocol (MCP) tools compatible with LangChain and LangGraph agents.Jul 2026
- mcporterTypeScript runtime and CLI for connecting to configured Model Context Protocol servers.Jul 2026
- mcp-proxyA MCP server which proxies requests to a remote MCP server over streamable HTTP or SSE.Jul 2026
- mcp-atlassianThe Model Context Protocol (MCP) Atlassian integration is an open-source implementation that bridges Atlassian products (Jira and Confluence) with AI language models following Anthropic's MCP specification.Jul 2026
- open-webuiOpen WebUIJul 2026
- rulesyncUnified AI rules management CLI tool that generates configuration files for various AI development toolsJul 2026
