Agent skill

Lego Rl Config

Compose, edit, refactor, and validate Lego-RL train/eval/infer .env configs and reusable scripts/templates modules.

Apache-2.0Auto-check: notesAI & LLM Engineering

Install Lego Rl Config

skills CLI

$ npx skills add LegoX/Lego-RL --skill lego-rl-config -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI

$ gh skill install LegoX/Lego-RL lego-rl-config --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy

$ git clone --depth 1 https://github.com/LegoX/Lego-RL.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/lego-rl-config .claude/skills/lego-rl-config && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Paste into Claude Code

Install the "lego-rl-config" agent skill from https://github.com/LegoX/Lego-RL/tree/main/.agents/skills/lego-rl-config into .claude/skills/lego-rl-config/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "lego-rl-config", then confirm the skill loads.

Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

$skill-installer

$skill-installer install https://github.com/LegoX/Lego-RL/tree/main/.agents/skills/lego-rl-config

Type this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.

skills CLI

$ npx skills add LegoX/Lego-RL --skill lego-rl-config -a codex

Project install goes to .agents/skills/; add -g for ~/.codex/skills/.

GitHub CLI

$ gh skill install LegoX/Lego-RL lego-rl-config --agent codex

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/LegoX/Lego-RL.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.agents/skills/lego-rl-config .agents/skills/lego-rl-config && rm -rf skills-src

Use ~/.agents/skills/ instead of .agents/skills for a personal install.

Codex skills documentation · loads skills from .agents/skills/

Paste into Codex

Install the "lego-rl-config" agent skill from https://github.com/LegoX/Lego-RL/tree/main/.agents/skills/lego-rl-config into .agents/skills/lego-rl-config/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "lego-rl-config", then confirm the skill loads.

Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add LegoX/Lego-RL --skill lego-rl-config -a cursor

Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.

GitHub CLI

$ gh skill install LegoX/Lego-RL lego-rl-config --agent cursor

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/LegoX/Lego-RL.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.agents/skills/lego-rl-config .cursor/skills/lego-rl-config && rm -rf skills-src

Use ~/.cursor/skills/ instead of .cursor/skills for a personal install.

Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/

Paste into Cursor

Install the "lego-rl-config" agent skill from https://github.com/LegoX/Lego-RL/tree/main/.agents/skills/lego-rl-config into .cursor/skills/lego-rl-config/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "lego-rl-config", then confirm the skill loads.

Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

gemini skills install

$ gemini skills install https://github.com/LegoX/Lego-RL.git --path .agents/skills/lego-rl-config

--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.

skills CLI

$ npx skills add LegoX/Lego-RL --skill lego-rl-config -a gemini-cli

Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.

GitHub CLI

$ gh skill install LegoX/Lego-RL lego-rl-config --agent gemini-cli

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/LegoX/Lego-RL.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.agents/skills/lego-rl-config .gemini/skills/lego-rl-config && rm -rf skills-src

Use ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.

Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/

Paste into Gemini CLI

Install the "lego-rl-config" agent skill from https://github.com/LegoX/Lego-RL/tree/main/.agents/skills/lego-rl-config into .gemini/skills/lego-rl-config/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "lego-rl-config", then confirm the skill loads.

Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

GitHub CLI

$ gh skill install LegoX/Lego-RL lego-rl-config

Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).

skills CLI

$ npx skills add LegoX/Lego-RL --skill lego-rl-config -a github-copilot

Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.

Manual copy

$ git clone --depth 1 https://github.com/LegoX/Lego-RL.git skills-src && mkdir -p .github/skills && cp -r skills-src/.agents/skills/lego-rl-config .github/skills/lego-rl-config && rm -rf skills-src

Use ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.

GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/

Paste into GitHub Copilot

Install the "lego-rl-config" agent skill from https://github.com/LegoX/Lego-RL/tree/main/.agents/skills/lego-rl-config into .github/skills/lego-rl-config/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "lego-rl-config", then confirm the skill loads.

GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add LegoX/Lego-RL --skill lego-rl-config -a opencode

OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.

GitHub CLI

$ gh skill install LegoX/Lego-RL lego-rl-config --agent opencode

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/LegoX/Lego-RL.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.agents/skills/lego-rl-config .opencode/skills/lego-rl-config && rm -rf skills-src

Use ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.

OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/

Paste into OpenCode

Install the "lego-rl-config" agent skill from https://github.com/LegoX/Lego-RL/tree/main/.agents/skills/lego-rl-config into .opencode/skills/lego-rl-config/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "lego-rl-config", then confirm the skill loads.

OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

Facts

Skill name: lego-rl-config
GitHub stars: 113
Token cost: ~2.1k tokens
SKILL.md length: 849 words
Files: 3 (incl. references)
Skills in repo: 11
Repo updated: Oct 8, 2026
First seen: Oct 7, 2026
Licence: Apache-2.0

At a glance

Compose, edit, refactor, and validate Lego-RL train/eval/infer .env configs and reusable scripts/templates modules.

Works in 5 steps: Resolve The Workload → Compose Template Modules → Generate Or Edit The Config → …
Codex is asked to generate an experiment config
SKILL.md covers Related Operational Skills, Start Here, Workflow and Refactor Rules, plus 1 more section
Calls bash; needs WANDB_API_KEY

What it does

Lego Rl Config is an agent skill from LegoX/Lego-RL. Compose, edit, refactor, and validate Lego-RL train/eval/infer .env configs and reusable scripts/templates modules. Use when Codex is asked to generate an experiment config, migrate legacy wrappers into configs, edit template modules, dry-run a train/eval/infer workload for config validation, or explain the runner/template/site-env contract. For Claude-style operational commands use the one-to-one Codex counterparts $rl-check, $rl-run, $rl-status, $rl-dashboard, and $rl-k8s-sandbox-install.

Its SKILL.md is about 2.1k tokens, which your agent loads only when the skill is triggered. The skill folder holds 4 other files, including reference files (for example `agents/openai.yaml` and `references/config-generation.md`).

It sits in AI & LLM Engineering, covering Container orchestration and Refactoring. It works with Kubernetes and vLLM. The repository describes itself as: Lego-RL: Harness-Native Reinforcement Learning for Coding Agents. The licence is Apache-2.0.

When your agent uses it

Codex is asked to generate an experiment config
Migrate legacy wrappers into configs
Edit template modules
Dry-run a train/eval/infer workload for config validation

Example prompts

“/lego-rl-config”

Requirements

Docker
A credential in WANDB_API_KEY

Workflow steps

5 steps, taken from the step headings in SKILL.md.

Resolve The Workload
Compose Template Modules
Generate Or Edit The Config
Validate Through The Runner
Report

What it can do on your machine

Read from SKILL.md and the folder at commit 0731c95. It shows what the files ask for, not the result of running them.

Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
- bash
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names these keys or tokens, usually read from environment variables:
- WANDB_API_KEY
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Lego Rl Config loads about 2.1k tokens when it runs, and up to ~4.3k if it reads all its reference files. Until then it costs about 128 tokens; SKILL.md has 849 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it: ~128
When it runs · the whole SKILL.md, loaded when a task matches: ~2.1k
With references · SKILL.md plus every file in references/, read only if the agent opens them: ~4.3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check: notes

The automated check noted patterns worth knowing about, such as sudo or a known installer.

NoteMentions a .env fileSKILL.md:3

r, and validate Lego-RL train/eval/infer .env configs and reusable scripts/templates modules. Use when Codex is asked to

NoteMentions a .env fileSKILL.md:9

single runner per workload plus small `.env` experiment configs composed from

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from LegoX/Lego-RL at commit 0731c95, republished under its Apache-2.0 licence (© LegoX). 849 words, ~2,052 tokens.

Download SKILL.mdSave it as .claude/skills/lego-rl-config/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.

name: lego-rl-config
description: Compose, edit, refactor, and validate Lego-RL train/eval/infer .env configs and reusable scripts/templates modules. Use when Codex is asked to generate an experiment config, migrate legacy wrappers into configs, edit template modules, dry-run a train/eval/infer workload for config validation, or explain the runner/template/site-env contract. For Claude-style operational commands use the one-to-one Codex counterparts $rl-check, $rl-run, $rl-status, $rl-dashboard, and $rl-k8s-sandbox-install.

Lego-RL Config

Use this skill for Lego-RL configuration work. The current design is a single runner per workload plus small .env experiment configs composed from reusable modules under scripts/templates.

This skill follows the Claude /rl:* plugin's layering rule:

Scripts own deterministic behavior. Skills own orchestration, judgement, and the final report.

Use this skill for config and template work. Use the one-to-one Codex counterparts for Claude plugin operations:

$rl-check for /rl:check
$rl-run for /rl:run
$rl-status for /rl:status
$rl-dashboard for /rl:dashboard
$rl-k8s-sandbox-install for /rl:k8s-sandbox-install

Start Here

Find the repository root: it contains scripts/train/train.sh, scripts/infer/infer.sh, scripts/eval/eval.sh, and scripts/templates/README.md.
Read scripts/templates/README.md first. It is the authoritative runner and template contract.
Read references/config-generation.md before creating or refactoring configs.
Inspect only the files relevant to the requested workload: scripts/<kind>/<kind>.sh, scripts/<kind>/_template.env, scripts/<kind>/configs/*.env, scripts/<kind>/lib/*.sh, and the selected scripts/templates/**.env modules.
Preserve user-owned run configs. Do not rewrite unrelated configs, logs, checkpoints, trial outputs, or cluster state.

Workflow

1. Resolve The Workload

Classify the request as train, infer, or eval.

train: verl policy training, sync/async mode, VeOmni/FSDP engine, TRAIN_FILES, VAL_FILES, NNODES, N_NODES_TRAIN, N_NODES_ROLLOUT.
infer: batch trajectory generation through utils/eval_swerebench_filtered.py, INDEX_FILE, optional INSTANCES_FILE, RESULTS_DIR, OUTPUT_INDEX, single-node vLLM serving knobs such as GEN_TP, GPUS_PER_NODE, and VLLM_PORT.
eval: Harbor-native scoring, exact MODEL_PATH, exactly one of DATASET_PATH or DATASET_NAME, local plain-vLLM serving, generated Harbor JobConfig, and harbor run.

If a config path is provided, infer the kind from scripts/<kind>/.... If only a bare name is provided, search scripts/{train,infer,eval}/configs/. Ask only when multiple plausible configs match.

2. Compose Template Modules

Templates live under scripts/templates/**.env, and a config chooses them with TEMPLATE_MODULES. The config is sourced first, then modules are sourced in order from scripts/templates. Template defaults should use:

bash

: "${VAR:=default}"

That means explicit config values are authoritative, while modules provide defaults and derived values.

Use the current module ownership model:

runtime/process.env: process-level env, sockets, NCCL/logging defaults, tokenizer/thread knobs, Ray ports, and Ray object store memory.
backend/k8s.env and backend/docker.env: Harbor backend selectors and backend defaults.
harbor/common.env: Harbor agent, trial, validation, retry, resource, verifier, and timeout defaults shared across workloads.
scaffold/{ohsdk,oh,cc,oc}.env: agent identity and runtime image defaults.
verl/common.env: shared train-side verl data/model/actor/rollout/ref/ algorithm/topology/log defaults.
verl/{async,sync}.env: train mode entrypoint/config and mode-specific defaults.
verl/{veomni,fsdp}.env: train model-engine-specific actor/ref/router-replay overrides.
infer/{vllm,common}.env: infer single-node vLLM serving plus infer rollout/data/output/log defaults.
eval/{common,vllm}.env: Harbor-native eval job/data/log defaults plus single-node plain-vLLM serving defaults.

Keep TEMPLATE_MODULES at the end of configs so module names and derived defaults can depend on earlier explicit settings.

3. Generate Or Edit The Config

Write generated configs to exactly one of:

scripts/train/configs/<name>.env
scripts/infer/configs/<name>.env
scripts/eval/configs/<name>.env

Use the workload skeleton as the starting point:

scripts/train/_template.env
scripts/infer/_template.env
scripts/eval/_template.env

Keep configs readable as experiment records: template selection first, identity, runtime, model, data/output, topology or serving, optional overrides, then TEMPLATE_MODULES. Keep generated configs small; do not copy every template default into the config.

Important current variable names:

Use EXP_NAME, not EXP_TAG.
Train uses TRAIN_FILES and VAL_FILES.
Infer uses INDEX_FILE for the parquet passed to --index.
Eval uses MODEL_PATH directly; old MODEL_PRESET-based eval templates are not part of the current runner contract.
Eval must set exactly one of DATASET_PATH or DATASET_NAME.

Show full SKILL.md (339 more words)Show less

4. Validate Through The Runner

Do not re-implement runner checks. Use the workload runner's dry-run path:

bash

bash scripts/<kind>/<kind>.sh --dry-run scripts/<kind>/configs/<config>.env

Dry-run sources the config and modules, validates required variables, initializes local runtime state, prints === Final Environment ===, prints the launch command block, then exits before Ray startup, vLLM startup, Harbor job writing, or training/eval/infer execution.

For static syntax checks, use the commands in scripts/templates/README.md, for example:

bash

bash -n scripts/train/train.sh scripts/train/lib/*.sh
bash -n scripts/infer/infer.sh scripts/infer/lib/*.sh scripts/templates/infer/*.env
bash -n scripts/eval/eval.sh scripts/eval/lib/*.sh scripts/templates/eval/*.env

If validation fails, report the exact fatal/error lines and adjust only the config or template layer that owns the value.

5. Report

Answer in Chinese unless the user asked otherwise. Include:

config path or template path created/changed
template modules used or introduced
key resolved axes: kind, backend, scaffold, model, and topology/serving; for train also mode and engine
validation commands run and their result
any manual values still needed, especially data paths, checkpoint/model paths, kubeconfig/backend/site values, registry/image/mount values, and multi-node host/rank values

Refactor Rules

Keep existing runners as the execution contract: scripts/train/train.sh, scripts/infer/infer.sh, and scripts/eval/eval.sh.
Move reusable defaults to scripts/templates, not scripts/lib.
Keep scripts/lib for executable shell helpers and workload orchestration.
Generated configs belong under the workload's configs/ directory, never under scripts/templates.
Site-specific paths, kubeconfigs, registries, mounts, Docker hosts, and secrets remain in site env, caller env, or explicit run configs when the site requires them. Do not bake them into shared templates.
Do not delete legacy monolithic scripts unless the user explicitly asks. When migrating one, preserve behavior with one generated config plus reusable modules, then validate with --dry-run.
Prefer the existing module tree over introducing new dimensions. Add a module only when an existing module has the wrong ownership boundary.

Guardrails

Do not launch training/eval/infer unless the user explicitly asks and confirms.
Do not kill processes, clear /dev/shm, run ray stop, delete logs, delete checkpoints, or mutate the cluster.
Do not SSH to worker nodes; for multi-node flows, print the commands the user must run on each node.
Do not claim a config is validated without runner output.
Do not hardcode secrets such as WANDB_API_KEY, kubeconfig contents, registry credentials, or personal tokens into templates or generated configs.

Files

SKILL.md and 2 other files (references) in .agents/skills/lego-rl-config of LegoX/Lego-RL.

SKILL.md
agents/openai.yaml
references/config-generation.md

Open the folder on GitHubat commit 0731c95

Compare with similar skills

Lego Rl Config next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Lego Rl Config compared with similar skills
Skill	Stars	Used in	Tokens	Auto-check	Licence	Repo updated
Lego Rl Config this skillLegoX/Lego-RL	113	—	~2.1k	Automated safety check: Notes	Apache-2.0	3 days ago
Vllm Deploy K8svllm-project/vllm-skills	102	—	~2k	Automated safety check: Pass	Apache-2.0	6 mo ago
LLM Inference Scalingsickn33/agentic-awesome-skills	47k	1 repos	~2.1k	Automated safety check: Pass	MIT	2 days ago
Coreweave Hello Worldjeremylongshore/tons-of-skills-marketplace	2.8k	—	~1.4k	Automated safety check: Pass	MIT	yesterday
LLM Inference ScalingBagelHole/DevOps-Security-Agent-Skills	1.2k	—	~2k	Automated safety check: Pass	MIT	4 mo ago
Dstack Prototypingdstackai/dstack	2.3k	—	~1.6k	Automated safety check: Pass	MPL-2.0	2 days ago

Similar skills

Vllm Deploy K8s
vllm-project/vllm-skills
Deploy vLLM to Kubernetes (K8s) with GPU support, health probes, and OpenAI-compatible API endpoint.
102 GitHub stars~2k tokensUpdated 6 mo ago
AI & LLM EngineeringAuto-check passed
LLM Inference Scaling
sickn33/agentic-awesome-skills
Auto-scale LLM inference clusters on Kubernetes using KEDA, custom GPU metrics, and horizontal pod autoscaling.
47k GitHub starsUsed in 1 repo~2.1k tokens
AI & LLM EngineeringAuto-check passed
Coreweave Hello World
jeremylongshore/tons-of-skills-marketplace
Deploy a GPU workload on CoreWeave with kubectl. An agent skill from jeremylongshore/tons-of-skills-marketplace.
2.8k GitHub stars~1.4k tokensUpdated yesterday
AI & LLM EngineeringAuto-check passed
LLM Inference Scaling
BagelHole/DevOps-Security-Agent-Skills
Auto-scale LLM inference clusters on Kubernetes using KEDA, custom GPU metrics, and horizontal pod autoscaling.
1.2k GitHub stars~2k tokensUpdated 4 mo ago
AI & LLM EngineeringAuto-check passed
Dstack Prototyping
dstackai/dstack
Use with the dstack skill for model-serving work when the image, serving command, resources, backend/fleet choice, or service behavior is not proven.
2.3k GitHub stars~1.6k tokensUpdated 2 days ago
AI & LLM EngineeringAuto-check passed
Software Design Review
atilladeniz/Kubeli
Analyzes code based on John Ousterhout's "A Philosophy of Software Design".
387 GitHub stars~5.3k tokensUpdated 5 days ago
DevelopmentAuto-check passed

More from LegoX/Lego-RL

All 11 skills in this repo

Check
LegoX/Lego-RL
Preflight a Lego-RL config: answer "is it safe to launch this run right now?".
113 GitHub stars~2.9k tokensUpdated 3 days ago
Auto-check passed
Dashboard
LegoX/Lego-RL
Bring up the Lego-RL training dashboard (webui/) on whatever machine you are on, adapting to that box's layout instead of assuming this repo's paths.
113 GitHub stars~4.2k tokensUpdated 3 days ago
Auto-check passed
Run
LegoX/Lego-RL
Preflight and launch a Lego-RL run (train, eval or infer). An agent skill from LegoX/Lego-RL.
113 GitHub stars~3k tokensUpdated 3 days ago
Auto-check passed
Status
LegoX/Lego-RL
Diagnose a Lego-RL run that is already in flight (or just finished): which run is alive, how far it has got, and whether its numbers are healthy.
113 GitHub stars~3k tokensUpdated 3 days ago
Auto-check passed
K8s Sandbox Install
LegoX/Lego-RL
Guided install / scale-out of a sandbox Kubernetes cluster for the Lego-RL k8s backend (kubeadm 1.32 + containerd + flannel + ImageVolume, optionally nydus / a shared registry / an isolated dockerd).
113 GitHub stars~2.9k tokensUpdated 3 days ago
Auto-check: warnings
Rl Check
LegoX/Lego-RL
One-to-one Codex counterpart for Claude /rl:check. An agent skill from LegoX/Lego-RL.
113 GitHub stars~338 tokensUpdated 3 days ago
Auto-check passed

Works with

Questions about Lego Rl Config

What does Lego Rl Config do?

Compose, edit, refactor, and validate Lego-RL train/eval/infer .env configs and reusable scripts/templates modules. Lego Rl Config is an agent skill from LegoX/Lego-RL.env configs and reusable scripts/templates modules.

When should I use Lego Rl Config?

Lego Rl Config fits situations like: Codex is asked to generate an experiment config; migrate legacy wrappers into configs; edit template modules; dry-run a train/eval/infer workload for config validation.

How do I install Lego Rl Config in Claude Code?

Run `npx skills add LegoX/Lego-RL --skill lego-rl-config -a claude-code`. Or copy the skill folder (.agents/skills/lego-rl-config in LegoX/Lego-RL) into .claude/skills/lego-rl-config in your project. Claude Code loads it when a task matches its description.

How do I install Lego Rl Config in Codex?

Run `npx skills add LegoX/Lego-RL --skill lego-rl-config -a codex`. Or copy the skill folder (.agents/skills/lego-rl-config in LegoX/Lego-RL) into .agents/skills/lego-rl-config in your project. Codex loads it when a task matches its description.

Can I use Lego Rl Config in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add LegoX/Lego-RL --skill lego-rl-config -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/lego-rl-config, .gemini/skills/lego-rl-config, .github/skills/lego-rl-config and .opencode/skills/lego-rl-config in your project.

What does Lego Rl Config need to run?

Going by SKILL.md and its folder, Lego Rl Config needs the command-line tools its instructions call (bash) and credentials named WANDB_API_KEY. Our summary lists: Docker; A credential in WANDB_API_KEY.

Does Lego Rl Config access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Lego Rl Config safe to install?

Our automated static check of SKILL.md found notes only (mentions a .env file), nothing it rates as a warning. It is not a guarantee. Review the folder before installing.

What licence does Lego Rl Config use?

Lego Rl Config is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Lego Rl Config use?

About 2.1k tokens (SKILL.md is roughly 8.2k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 2.2k tokens, read only when the agent opens those files.

What are the alternatives to Lego Rl Config?

Skills that share tags, products or a category with Lego Rl Config: Vllm Deploy K8s (vllm-project/vllm-skills, 102 stars), LLM Inference Scaling (sickn33/agentic-awesome-skills, 47k stars), Coreweave Hello World (jeremylongshore/tons-of-skills-marketplace, 2.8k stars) and LLM Inference Scaling (BagelHole/DevOps-Security-Agent-Skills, 1.2k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Lego Rl Config?

LegoX (a GitHub organization) maintains it in LegoX/Lego-RL, which has 113 GitHub stars. The repository holds 11 skills in this directory. The repository was last updated on October 8, 2026.

Source: LegoX/Lego-RL on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.

Install Lego Rl Config

Facts

At a glance

What it does

When your agent uses it

Example prompts

Requirements

Workflow steps

What it can do on your machine

Tool permissions

Runs code

Network

Credentials

Context cost

Safety

SKILL.md

Lego-RL Config

Related Operational Skills

Start Here

Workflow

1. Resolve The Workload

2. Compose Template Modules

3. Generate Or Edit The Config

4. Validate Through The Runner

5. Report

Refactor Rules

Guardrails

Files

Compare with similar skills

Similar skills

More from LegoX/Lego-RL

Related

Questions about Lego Rl Config