Komodo

v1.0.0

Interact with Komodo Core API using this project. Use when the user wants to list, manage, deploy, or execute operations against Komodo resources (servers, s...

0· 256·0 current·0 all-time
byAlwaysBeDream@onekill0503

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for onekill0503/komodo-skill.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "Komodo" (onekill0503/komodo-skill) from ClawHub.
Skill page: https://clawhub.ai/onekill0503/komodo-skill
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install komodo-skill

ClawHub CLI

Package manager switcher

npx clawhub@latest install komodo-skill
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Suspicious
medium confidence
Purpose & Capability
Name/description match the implementation: the code and scripts call a Komodo client and provide list/create/update/deploy/run/logs operations. However the registry metadata declared no required environment variables while the code (openclaw.ts) and SKILL.md require KOMODO_URL, KOMODO_API_KEY, and KOMODO_API_SECRET — that mismatch is unexpected and misleading.
Instruction Scope
SKILL.md limits runtime actions to calling the Komodo API and running bundled scripts. The instructions do not ask the agent to read unrelated local files or external endpoints outside the Komodo API. Note: the bundled komodo_client exposes terminal/exec and websocket helpers (remote command execution on containers/servers) — this is consistent with the skill's stated purpose but is high-privilege behavior toward the Komodo-managed infrastructure.
Install Mechanism
There is no install spec; package.json lists a dependency on komodo_client and scripts are prebuilt (scripts/*.js include the bundled komodo_client code). No external downloads or obscure URLs are used by the skill bundle itself.
!
Credentials
The skill legitimately requires KOMODO_URL, KOMODO_API_KEY, and KOMODO_API_SECRET to talk to Komodo Core, and the SKILL.md and openclaw.ts make that explicit — but the registry metadata claims no required env vars. That omission could cause users to enable this skill without realizing it needs sensitive credentials. No other unrelated credentials or config paths are requested.
Persistence & Privilege
always is false and the skill doesn't request system-wide privileges or attempt to modify other skills. It can perform destructive operations (destroy stacks/deployments, execute remote commands) on the Komodo-managed infrastructure, which is expected for this tool but is operationally powerful.
What to consider before installing
This skill appears to implement what it claims (a Komodo management toolkit) but the registry metadata incorrectly lists no required environment variables. Before installing: 1) Only provide KOMODO_URL, KOMODO_API_KEY, and KOMODO_API_SECRET if you trust the skill and the Komodo instance; the key/secret are powerful — prefer a least-privilege key and rotate it after testing. 2) Understand that the bundled client supports remote terminal/exec and can start/stop/destroy resources; run the skill in an isolated environment and avoid granting it more than the Komodo credentials it needs. 3) If you rely on the registry metadata for automated checks, update or confirm the skill entry to include the three env vars so you are not surprised at runtime. 4) Review the scripts (they are included) and, if you want to restrict risk, run them manually (node scripts/*.js) rather than enabling autonomous invocation for an agent. 5) If you need higher assurance, ask the publisher for an explicit manifest of required env vars and a rationale for any destructive operations the skill performs.

Like a lobster shell, security has layers — review code before you run it.

latestvk97bthqdzmbc7whzeavzscy2vn82fk9d
256downloads
0stars
1versions
Updated 1mo ago
v1.0.0
MIT-0

You are an expert at managing Komodo infrastructure using this project. The openclaw.ts module exports a pre-configured komodo client authenticated via API key and secret from environment variables. All scripts in scripts/ import from it.

Environment Variables

VariableDescription
KOMODO_URLBase URL of Komodo Core (e.g. https://komodo.example.com)
KOMODO_API_KEYAPI key
KOMODO_API_SECRETAPI secret

Available Scripts

Every script can be run two ways — pick whichever fits the environment:

node scripts/<script>.js [args]        # compiled JS, no tooling needed
bun run scripts/<script>.ts [args]     # TypeScript source, requires Bun

list — Inspect resources

node scripts/list.js <type>
TypeOutput
serversState, address, region, periphery version, CPU, memory, disk, load average
stacksState, services & images, repo/branch, deployed vs latest commit
deploymentsState, image, update availability, attached build
buildsState, version, last built, repo/branch, built vs latest commit
reposState, last pulled/built, repo/branch, cloned/built/latest commit
proceduresState, stage count, last run, next scheduled run
actionsState, last run, next scheduled run

create — Create a resource

node scripts/create.js <type> <name> [json-config]

Config is optional — omit to use Komodo defaults.

node scripts/create.js stack my-stack '{"repo":"org/repo","branch":"main"}'
node scripts/create.js deployment my-dep '{"image":"nginx:latest","server_id":"<id>"}'
node scripts/create.js build my-build '{"repo":"org/repo","branch":"main"}'
node scripts/create.js repo my-repo '{"repo":"org/repo","branch":"main","server_id":"<id>"}'
node scripts/create.js procedure my-proc
node scripts/create.js action my-action '{"run_at_startup":false}'

Types: stack deployment build repo procedure action


update — Patch resource config

Applies a partial JSON merge — only specified fields change, everything else stays.

node scripts/update.js <type> <name> '<json>'
node scripts/update.js stack my-stack '{"branch":"main"}'
node scripts/update.js deployment my-dep '{"image":"nginx:1.27"}'
node scripts/update.js build my-build '{"version":{"major":2,"minor":0,"patch":0}}'
node scripts/update.js repo my-repo '{"branch":"develop"}'
node scripts/update.js procedure my-proc '{"schedule":"0 0 * * *","schedule_enabled":true}'
node scripts/update.js action my-action '{"run_at_startup":false}'

Types: stack deployment build repo procedure action


run — Execute and wait for completion

Runs a procedure, action, or build and blocks until done. Prints update ID, status, duration, and logs.

node scripts/run.js <type> <name>
node scripts/run.js procedure my-proc
node scripts/run.js action my-action
node scripts/run.js build my-build

deploy-stack — Deploy a stack

Deploys a stack and waits for completion. Prints the full update result and logs.

node scripts/deploy-stack.js <stack-name>

stack-ctrl — Control a stack

node scripts/stack-ctrl.js <action> <stack-name>

Actions: start stop restart pull destroy


deployment-ctrl — Control a deployment

node scripts/deployment-ctrl.js <action> <deployment-name>

Actions: deploy start stop restart pull destroy


get-logs — Fetch logs

node scripts/get-logs.js stack <stack-name> [tail]
node scripts/get-logs.js deployment <deployment-name> [tail]
node scripts/get-logs.js container <server-name> <container-name> [tail]

Writing New Scripts

When adding a new script, follow this pattern:

import { komodo } from "../openclaw.ts";
import { printUpdate } from "./print-update.ts"; // for execute_and_poll results

// Read
const items = await komodo.read("ListStacks", {});

// Write (create/update/delete)
await komodo.write("CreateVariable", { name: "KEY", value: "val" });

// Execute and wait for completion
const result = await komodo.execute_and_poll("DeployStack", { stack: "name" });
printUpdate(result);

Key rules:

  • Always import komodo from "../openclaw.ts" — never initialize the client directly
  • Always use execute_and_poll (not execute) when waiting for a task to finish
  • Use printUpdate(result) from ./print-update.ts to display execution results
  • Read args from process.argv — never hardcode resource names or credentials
  • After writing a new .ts script, rebuild with bun run build to produce the .js counterpart

Instructions

When the user provides a request via $ARGUMENTS:

  1. Identify the resource type and operation needed.
  2. Use an existing script first — check the list above before writing new code.
  3. If adapting an existing script, show the exact command to run.
  4. If writing a new script, place it in scripts/, follow the pattern above, and remind the user to rebuild.
  5. Always show both run options: node scripts/<script>.js and bun run scripts/<script>.ts.
  6. For execution results (deploy, run, ctrl), the output always includes: update ID, status, success, duration, operator, and per-stage logs.

The user's request: $ARGUMENTS

Comments

Loading comments...