ohmg

Ultimate multi-agent framework for Google Antigravity. Orchestrates specialized domain agents (PM, Frontend, Backend, Mobile, QA, Debug) via Serena Memory.

Install with Tessl CLI

npx tessl i github:supercent-io/skills-template --skill ohmg

What are skills?

6.40x

Quality

41%

Does it follow best practices?

Impact

96%

6.40x

Average score across 3 eval scenarios

Optimize this skill with Tessl

npx tessl skill review --optimize ./.agent-skills/ohmg/SKILL.md

SKILL.md

Review

Evals

Evaluation results

90%

86%

Onboarding Script for Multi-Agent Development Environment

Project setup and CLI configuration

Criteria

Without context

With context

Bun for install

100%

Doctor verification

33%

100%

Config file location

100%

agent_cli_mapping key

100%

Frontend mapped to gemini

100%

Backend mapped to codex

100%

PM mapped to claude

100%

QA mapped to claude

100%

uv prerequisite

Skills directory reference

100%

Without context: $0.7105 · 2m 11s · 30 turns · 31 in / 7,053 out tokens

With context: $0.3348 · 1m · 21 turns · 208 in / 3,358 out tokens

100%

75%

Automating a Feature Development Cycle with Specialized Agents

Agent spawning and workflow orchestration

Criteria

Without context

With context

spawn command prefix

100%

PM agent used

100%

Backend agent used

100%

Frontend agent used

100%

QA agent used

100%

Task quoted string

100%

Session ID present

33%

100%

Sequential ordering

100%

No wrong agent types

100%

Without context: $0.1370 · 41s · 11 turns · 16 in / 2,167 out tokens

With context: $0.2144 · 40s · 13 turns · 262 in / 2,050 out tokens

100%

82%

Real-Time Observability Tooling for Multi-Agent Sessions

Dashboard monitoring and Serena Memory

Criteria

Without context

With context

Terminal dashboard command

100%

Web dashboard command

100%

Correct web port

100%

Serena Memory path

100%

State file listing

30%

100%

URL output

100%

No wrong dashboard commands

100%

Without context: $2.8448 · 7m 9s · 94 turns · 2,089 in / 23,061 out tokens

With context: $0.2746 · 53s · 17 turns · 202 in / 2,819 out tokens

Evaluated: 3 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Onboarding Script for Multi-Agent Development Environment Automating a Feature Development Cycle with Specialized Agents Real-Time Observability Tooling for Multi-Agent Sessions

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.