Video plannedVideo planned
Watch and steer a supervised run
Show dispatch, slot assignment, live terminal output, decision gates, and how the human can redirect before trusting the result.
Active development previewFarmslot is moving quickly while the product is finalized in one repo. Experimental features, large/transient files, APIs, runner contracts, fixtures, and docs may change while the project matures.
Why it matters
Scale your own workflow without losing trust.
Farmslot is the product layer around agents: a framework for many project types, an OS for coordinating the fleet, an IDE for watching and steering work, and a proof-first recipe layer for visually and mechanically validating the result.
Agentic engineering OS
Coordinate projects, machines, slots, runners, hooks, run state, artifacts, and memory as one operating environment.
Open docs →Watch-and-steer IDE
Use Command Center as the cockpit: live terminals, decisions, app state, artifacts, and intervention while the run is still moving.
Open docs →Proof-first trust layer
Validate visually and mechanically with recipes at the core: real actions, assertions, screenshots, traces, HUD intent, and proof artifacts.
Open docs →Project-type framework
Bring mobile apps, web apps, extensions, CLIs, services, desktop apps, and headless systems into the same operating loop.
Open docs →Runner capacity leverage
Turn multiple model runners, computers, and subscriptions into coordinated parallel engineering capacity.
Open docs →Mobile supervision
Keep approvals, workers, runs, and terminals visible when you step away from the workstation.
Open docs →Recursive improvement
Turn failures, flakes, and review comments into better prompts, docs, recipes, checks, and adapters for the next run.
Open docs →Demo plan
Each short clip should explain one product promise.
The best demos are narrated feature explainers: dispatch real work, watch and steer it live, validate with recipes, review the evidence, and feed lessons back into the next run.
Video planned
Video plannedVideo planned
Recipe evidence run
Show a recipe run turning setup, real actions, visual validation, assertions, HUD messages, and proof artifacts into review evidence.
Video plannedVideo planned
Companion mobile supervision
Show the operator checking workers, terminal status, and run progress from the mobile companion without losing control.
Recording checklist
The landing page currently uses static public-safe placeholders. When final clips are recorded, keep them short, caption-friendly, poster-backed, and verify representative frames before publishing.
View recording plan →Recursive feedback
Every run should make the next run better.
Farmslot treats validation and retrospectives as part of the product loop. Runs create evidence, review finds gaps, retrospectives capture durable lessons, and those lessons improve recipes, prompts, docs, and adapters.
Read the feedback loop- RunExecute in an isolated slot with live logs, artifacts, and human decision gates.
- ObserveCapture terminal streams, screenshots, traces, summaries, and reviewer notes.
- ReviewUse humans and independent model reviewers to find gaps before merging.
- RetrospectConvert failures, flakes, and repeated review comments into curated learnings.
- CurateKeep only reusable lessons that can become checks, prompts, recipes, docs, or adapters.
- ImproveUpdate prompts, recipes, docs, and project adapters before the next run.
Start small, then deepen the contract.
Import a project with hooks, add a runner manifest, adopt recipes where behavior needs proof, then use retrospectives to keep improving the system.