Get This Tool
Agentic FC
Pricing
- Model
- Free
Summary
Most AI agent experiments stall at the toy stage because there is no persistent world with real stakes — no season to lose, no squad to manage, no match clock ticking. Agentic FC gives your agent a full football club to run, right now, in a terminal.
Agentic FC is a Go-based, MIT-licensed football management simulation where AI agents control a club through MCP tool calls — reading match state, setting tactics, reacting to news — while a human watches through a terminal TUI with ASCII match scenes and live commentary. The engine is deterministic and seeded, so simulations replay identically, which matters when you are debugging agent decision loops rather than blaming random variance. The agent shapes the in-game Manager's mindset rather than clicking menus, making this a concrete testbed for studying how an LLM actually behaves inside a continuous decision loop. The project has two stars on GitHub and zero open pull requests — the community is early. Teams pushing beyond the built-in MCP tool surface will be writing Go extensions against a codebase that is still accumulating commits.
Bottom line: Pick this to validate whether your MCP agent can hold a coherent strategy across a full match cycle; plan a different architecture when you need multi-club leagues, persistent season databases, or a tool surface that someone else has battle-tested.
Community Performance Report Card
No community ratings yet. Be the first to rate this tool!
Community Benchmarks Community
Sign in to submit a benchmarkNo community benchmarks yet. Be the first to share a real-world data point.
Pros
Sign in to edit- Deterministic seeded simulation engine, so you can replay the exact same season with a different agent strategy and know any behavioral difference is the agent's fault — not a random dice roll.
- MCP-native agent control loop, meaning your agent stays in a continuous read-decide-observe cycle across match events rather than firing a single prompt and waiting, which makes multi-step reasoning failures visible and debuggable.
- Terminal TUI with ASCII match scenes and commentary, so you can watch agent decisions play out in real time without building a frontend — useful for demos and for catching the moment an agent tactically loses the plot.
- MIT-licensed Go source with self-hosting support, so you run the simulation on your own hardware, keep all agent traces private, and modify the engine without asking anyone for access.
- AGENTS.md and CLAUDE.md contributor files included, which means the project explicitly documents how agents are expected to interact with the codebase — reducing the guesswork when wiring up a new agent client.
Cons
Sign in to edit- The MCP tool surface covers what the README describes — Manager mindset control and match observation — but multi-club league management, a transfer market, and deep squad statistics are not documented as available endpoints; teams whose research question depends on those features hit a wall before their first experiment and are writing Go extensions against early-stage source.
- The project has two GitHub stars and no open pull requests at the time of scraping, which means bug reports go unanswered until the maintainer picks them up, documentation gaps stay gaps, and any integration breakage after a Go dependency update is yours to diagnose alone.
- Teams that need a proven, high-traffic agent sandbox with existing community tooling — scenario libraries, recorded agent runs, public leaderboards — will abandon this for a dedicated research environment like OpenSpiel or a custom gym, because Agentic FC's community does not yet provide those resources.
Community Reviews
Sign in to write a reviewNo reviews yet. Be the first to share your experience.
About
- Platforms
- Terminal / TUI, Go runtime
- API Available
- Yes
- Self-Hosted
- Yes
- Last Updated
- 2026-07-09T18:55:50.926Z
Best For
Who it's for
- Developers building AI sports agents
- Terminal-based simulation enthusiasts
- MCP tool integration experiments
- Deterministic game engine research
What it does well
- AI agents managing football clubs via MCP
- Watching live matches in terminal TUI
- Deterministic seeded world simulations
- Studying agent-driven decision making in sports sims
Integrations
Discussion Community
Sign in to commentNo discussion yet. Sign in to start the conversation.
Compare Agentic FC
Spotted incorrect or missing data? Join our community of contributors.
Sign Up to ContributeCommunity Notes & Tips Community
Sign in to contributeBe the first to contribute. General notes, observations, gotchas, and tips from people who use this tool day-to-day.
Frequently Asked Questions
- Is Agentic FC free?
- Yes — Agentic FC is fully free to use. There is no paid tier.
- Is Agentic FC open source?
- Yes. Agentic FC is open source.
- Does Agentic FC have an API?
- Yes. Agentic FC exposes a developer API. See the official documentation at https://github.com/gaemi/agentic-fc for details.
- Can I self-host Agentic FC?
- Yes. Agentic FC supports self-hosting on your own infrastructure.
- What platforms does Agentic FC support?
- Agentic FC is available on: Terminal / TUI, Go runtime.
Hours Saved & ROI Stories Community
Sign in to contributeBe the first to contribute. Concrete time/cost savings, with context. e.g. "Cut my code review backlog from 4h to 45m per week."
Curated lists that include this category
Agentic FC is an open-source football management simulation written in Go where the player is not a human clicking menus — the player is an AI agent issuing MCP tool calls. The agent reads the world state, sets the Manager’s mindset, and watches probabilistic football decisions cascade into match events. Humans observe the result through a terminal TUI that renders ASCII match scenes and running commentary. The world engine is seeded and deterministic, so the same seed produces the same season, making it possible to isolate whether a different agent strategy — not a different random roll — changed the outcome.
The core differentiating feature is the MCP-native control loop. Instead of wrapping a simulation in a one-shot prompt, the agent operates in a continuous read-decide-observe cycle through MCP tool calls, giving it the same kind of persistent context pressure a real management decision loop would impose. This makes the project useful not just as a game but as a repeatable environment for studying how an LLM agent degrades, fixates, or adapts across a long time horizon — something a single API call cannot reveal.
The tool fits developers who want a structured, non-trivial environment for MCP agent experiments without building their own simulation engine from scratch. It breaks down when the research question requires features the current tool surface does not expose: multi-club competition, transfer markets, or deep squad statistics. The project is at an early commit count with a small community, so gaps in documentation and missing tool endpoints are filled by reading Go source, not by searching Stack Overflow.