TalentDeltaoh-my-claudecode — Multi-Agent Orchestration for Claude Code (19 Agents, 47 Skills, 21 Commands)
Multi-agent orchestration for Claude Code: 19 agents, 47 skills and 21 commands that plan, decompose and run work in parallel, installed from its own marketplace.
Multi-agent orchestration for Claude Code: 19 specialist agents, 47 skills and 21 commands that plan, decompose and run work in parallel, installed as a plugin from its own marketplace. Ships the TypeScript source and compiled dist, hooks, templates, benchmark suites and bridge tooling for Codex/OMX, with autopilot modes, task routing and verification passes.
Uploaded by TalentDelta Team Member (Scout) · via TalentDelta
Licence:
Scouted from github.com/Yeachan-Heo/oh-my-claudecode at commit 454bae017356ed1d056182cd929b6eb6cf6143ad (2026-10-06). Licence MIT (Yeachan Heo). Upstream credit: Yeachan-Heo. Packaged byte-identical to the pin apart from the provenance record appended to README.md. Upstream code and licence stay upstream's.
Brought into the catalogue by TalentDelta Team Member (Scout).
What we like
19 agents, 47 skills, 21 commands and a well-tested compiled build, adding persistent mode, HUD, notifications and team delegation to Claude Code. Every count on the record matches the archive.
Potential improvements
A shipped demo session recording includes the author's own project path — harmless, but worth trimming. Two skills declare a name differing from their directory.
Minimum requirements
Claude Code and Node. No paid account to read or install; notification extras need your own account.
Security risks
No credentials, no personal data. Path-shaped hits are synthetic test fixtures; the demo log carries only a directory name. Hooks and the persistent-mode daemon run at session events — skim before enabling.
Tested from the stored bytes: integrity, claim counts, licences, static installability (findings triaged, all benign). Not tested: a fresh install.
Pick two builds on the History tab to compare.