← Back
Home › Systems › Frontier AI research

rl-environments

RL environment platform for agentic training

coding and computer-use gyms

SystemTechniques · 31Primitives · 99 rl-environments Model context protocolagentic-ai · medium: JSON-RPC protocol exposing tools, resources and prompts from servers to LLM clients, with capability negotiationUser simulationagentic-ai · medium: LLM-driven simulated users following scripted goals to test conversational agentsDeterministic simulation testingdistributed · medium: Run the whole system in one thread under a seeded scheduler with simulated network, disk and clock, so every failure replays from its seed.›Durable job queuedistributed · medium: Persisted tasks that workers lease for a visibility timeout, ack on success and retry with backoff; at-least-once execution that survives worker and queue crashes.›Primary-backup replicationdistributed · medium: One node orders all writes and ships them to standbys; the acknowledgement rule (none, one, all) sets what a crash loses, and an external arbiter decides who takes over.›Vm snapshot restorehardware-os · medium: Snapshot an initialised VM/isolate heap and restore it for fast cold startsBenchmark contamination detectionmachine-learning · medium: Detect test-set leakage into training data via n-gram overlap, perplexity and canary checksRecord replaydistributed · low: Record every nondeterministic input (time, randomness, I/O, model and tool outputs, per-tick player input) and re-execute deterministically for debugging, replays and state rebuildAccessibility treeagentic-ai · low: Extracting the browser/OS accessibility tree (roles, names, bounds) as a compact text view of a UI for agentsAgent tracingagentic-ai · unknown: Browser automation protocolagentic-ai · low: CDP / WebDriver BiDi commands and events over websockets to drive browsersInput event synthesisagentic-ai · low: Inject synthetic mouse, keyboard and touch events at OS or browser level (CGEvent, SendInput, uinput, CDP Input) with realistic timingScreen captureagentic-ai · low: Damage-region screen capture with GPU surfacesTool callingagentic-ai · low: The model emits a structured call (id, name, JSON arguments) instead of prose; your code validates, runs it and returns a result tagged with the same id.›Containers and Kubernetescloud · low: Containers package a service with its dependencies; Kubernetes schedules them across machines. Use the smallest orchestrator that does the job.›Hdr histogramdata-structures · low: Log-linear bucketed histograms with bounded relative error for latency percentilesProcess supervisionhardware-os · low: A parent process starts services, reaps their exits and restarts them under a delay and burst-limit policy, then stops them with SIGTERM before SIGKILL.›Mutual TLSsecurity · low: TLS where the client also proves a certificate key, so every connection carries a verified, rotatable machine identity.›Graceful drainingsystems · low: On a termination signal, stop taking new work, finish in-flight work until a deadline, tell clients to reconnect elsewhere, then exit; clients resume from a token.›Json rpcnetworking · low: JSON-RPC 2.0 requests, notifications and batchingServer-sent eventsnetworking · low: One long-lived HTTP response carries text events one way; each event may carry an id, so a reconnecting client names its last one and the server resumes.›JSON Schema validationsecurity · low: Compile a JSON Schema once, then check every untrusted document against it, within hard size and depth limits, and report failures as JSON Pointers.›OAuth 2 and OpenID Connectsecurity · low: Delegated authorization (OAuth 2) and login on top of it (OIDC); use authorization code + PKCE for every user-facing client.›+ 24 more techniques+ 83 more primitives

Similar systems

Share the most building blocks with this one.

agent-browser-isolation 46 %browser-test-grid 32 %computer-use-agent 31 %serverless 30 %ci-runner-fleet 28 %