Agents Orchestrator Skill — Security skill for Claude Code
Orchestra: a cross-vendor execute → audit → escalate loop for CLI coding agents (Claude Code, Codex, Antigravity, +15).
How to install Agents Orchestrator Skill
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open Soliman-Elhassanein/Agents-Orchestrator-Skill and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Agents Orchestrator Skill does
Orchestra: a cross-vendor execute → audit → escalate loop for CLI coding agents (Claude Code, Codex, Antigravity, +15). Balances work across your subscriptions' quotas. Built on amElnagdy/delegate-skills.
Alternatives in Security
- Octoboss Clean Contexts — You are the Octoboss — a cross-tentacle orchestrator 1.4k ★
- Pocketpaw — Your AI agent in 30 seconds 775 ★
- Engage.Exploit — Execute Phase 4 - Exploitation and Access Establishment 348 ★
README
Orchestra
A multi-agent coding loop that works across vendors: **execute → audit → escalate**. You talk to one **Communicator** (whatever CLI you like). It queues your requests and hands each one to an **Execution Orchestrator**, which splits it across parallel **Executors**. An **Audit Orchestrator** has parallel **Auditors** review the result and decides a score. The result is accepted only when the score passes your threshold **and** the hard checks pass: gates green, no open blocker, every subtask landed, every auditor done. Otherwise the findings go back to the executors, and on the cycles you choose, stronger models take over.
**The goal: get the most out of the subscriptions you already pay for.**
- Strong models orchestrate; cheap models do the bulk of the work.
- Every assignment is weighted by how fast each subscription's 5-hour and weekly allowance is being used, so they all drain at an even pace and none runs dry early.
- Nothing spends tokens while it's waiting.
Any role can be any of 18 CLIs: Claude Code, Codex, Antigravity, OpenCode, Cursor, Copilot, Grok, Kimi, Qoder, Aider, Cline, Pi, Oh My Pi, Muse, Vibe, Warp, ZCode, Command Code. For example, a Codex orchestrator can drive Claude and Antigravity executors.
*Despite the repository's name, Orchestra isn't a Claude Code skill: it's a small Node CLI you install into each project.*
you ◄──► Communicator ──► queue ──► Execution Orchestrator ──► Executors ×N (dependency waves, own worktrees)
▲ ▲ ▲ ▲ ▲ │ conflicts go back to the orchestrator
│ │ │ questions │ │ findings + digests of ▼ merged → gates on that commit
│ │ │ (one batch) │ └── failed attempts ─── Audit Orchestrator ──► Auditors ×M
│ │ └────────────────────┴───────────────────────────────┤ (informed + fresh, read-only)
│ └──── accepted: score ≥ threshold, gates pass, no blocker ┘
Related Skills
Agent Advisor
Hub for selective AI-agent routing across Codex, Claude Code & Antigravity: auditable solo/delegate/audit/full
LLM Spend Audit
Your agent setup charges rent on every session, before it does any work. Three stdlib-only instruments: what y
Security Health Check
Execute the security-health-inline skill for inline orchestration. You ARE the orchestrator. Do not spawn a se
Intutic
The open source circuit breaker for AI agents. Real-time security, secret DLP, graph guardrails and loop burn
Safeai
🛡️ Local security gateway & Human-in-the-Loop firewall for AI coding agents (Copilot, Cursor, Antigravity, Cl
Mcpick
Vendor-neutral MCP configuration manager — one CLI to add, toggle, and audit MCP servers and skills across eve
Related Agents
Paf Reviewer
Runs one phone-a-friend relay or review command in the background and reports back a receipt plus the findings
Security Sentinel
Deep security review focused on OWASP Top 10, injection, auth/authz, data exposure, and dependency vulnerabili
Verify
Claude Code runner adapter. 격리 snapshot에서 Codex/Antigravity/Tailscale Aperture reviewer를 조합해 교차검증. 보안 결정권 없음,