marin-community/MarinSkyRL

rl-supervisor-init

Initialize and run a focused RL operations session across Iris and explicitly selected supported backends: inventory jobs, dispatch diagnosis, perform authorized lifecycle actions, and maintain a reproducible handoff for an explicit list of experiment…

原文语言：英语

commit

Lint, run the pre-PR checks, commit, push, and author or update the branch's pull request in the required plain-text format. Use when committing, pushing, or creating/updating a PR.

原文语言：英语

consult-echo

Search and cite Marin's shared Echo wiki, repository index, issues, pull requests, and optional Discord history. Use when prior cross-project decisions, incidents, workflows, exact errors, or reusable knowledge could inform a task, and before adding or…

原文语言：英语

debug

Debug code bugs and operational faults while recording distilled hypotheses, evidence, and outcomes in Marin's shared Echo work log.

原文语言：英语

file-issue

File a GitHub issue for a bug or improvement found this session.

原文语言：英语

task-logbook

Maintain a cross-project task or research log in Echo. Use for multi-step investigations, experiments, long-running implementation work, or any task whose milestones must survive the current agent session without adding a repository logbook file.

原文语言：英语

write-design-doc

Publish or revise a concise cross-project design proposal in Echo. Use for an explicit design task or a design-level change whose context, decision, interfaces, tradeoffs, and rollout need review without adding a repository design document.

原文语言：英语

write-ops-log

Publish a durable debugging or infrastructure incident record to Echo. Use when a multi-step investigation reaches a conclusion, a failure needs a postmortem, or a recovery pattern should be visible across Marin projects.

原文语言：英语

write-tests

Write or revise tests with an emphasis on behavior, regression coverage, pytest style, and avoiding "slop tests." Use when adding tests, fixing failing tests, reviewing test quality, or deciding what test would catch a bug.

原文语言：英语

writing-style

Marin house writing style. Use when drafting or revising Marin-authored prose, including commit messages and GitHub PR, issue, or comment text.

原文语言：英语

rl-job-health-deep-dive

Diagnose one live or suspect RL job on Iris or an explicitly selected supported backend and return an evidence-backed KILL, NO-KILL, or ERROR recommendation. Use when ordinary state and metric checks cannot distinguish progress from a silent stall, or when…

原文语言：英语

build-gpu-rl-image-iris

Build and validate compatibility MarinSkyRL GPU-RL containers when the user explicitly requests one. Maintained Iris training and utilities use frozen environments instead.

原文语言：英语

rl-agentic-launch-iris

Validate, submit, and observe agentic MarinSkyRL training on Iris when the configuration uses Harbor, Daytona, terminal-bench, or another sandboxed agent harness.

原文语言：英语

rl-standard-launch-iris

Validate, submit, and observe standard MarinSkyRL training on Iris for dataset-backed rewards without Harbor, Daytona, terminal-bench, or another agent harness.

原文语言：英语

rl-agentic-job-cleanup

Preserve and, when authorized, publish a terminal agentic Iris RL run: stop pending retries, select and validate a checkpoint, export model weights, preserve metrics and traces, verify destinations, and only then reclaim storage. Use for runs with an agent…

原文语言：英语

experiment-status-artifact

Render a fleet of experiment runs as a compact, operational status artifact. Use when a user needs a dashboard or overview across multiple runs, arms, datasets, or configurations rather than a prose report.

原文语言：英语

rl-standard-job-cleanup

Preserve, validate, and hand off a terminal standard MarinSkyRL run backed by dataset rows and programmatic rewards, without an agent harness or per-trial artifacts. Use after completion or failure, or before an authorized cancellation. Use…

原文语言：英语