一键在 Manus 中运行任何 Skill

apeiron

星标1

分支0

更新时间2026年7月17日 10:19

The ONE-COMMAND orchestrator that wires the full delegate-team arsenal for any task in any project. Use when the user says "use everything you got", "use full engineering arsenal", "go all-in", "/Apeiron <task>", "/ship <task>", "use all capabilities", "full stack", "end to end", "everything intelligently". Composes 38 curated integrations (superpowers, Waza, unslop, autoresearch, agent-kernel, ...) and routes tasks through 7-stage autopilot chain (PREWARM → BRAINSTORM → PLAN → EXECUTE → REVIEW → QUALITY-GUARD → REPORT). Auto-discovers 1890+ skills across 3 sources (apeiron, agents, claude) with intelligent dedupe. Includes apeiron-uni — the smart universal wrapper that detects the calling runtime (Apeiron / codex / claude / gemini / opencode / mmx / shell) and dispatches the right flow. Skip this skill only when the user wants a SPECIFIC tool (e.g. "run /autoresearch with metric X" or "delegate to aonios-agent").

安装

用 Codex 或 Claude 帮你安装复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。

在 Manus 中运行

来源

imMamdouhaboammar

imMamdouhaboammar/delegate-team

打开 GitHub 仓库查看创作者相关仓库

下载

在 Manus 中运行

相关职业SOC

基于 SOC 职业分类

其他计算机职业计算机与数学类职业·SOC 15-1299

文件资源管理器

5 个文件

SKILL.md

readonly

同仓库更多 Skills

同仓库

delegate-team

imMamdouhaboammar/delegate-team

Complete agentic engineering supersystem — one command orchestrates the entire engineering chain. Use when the user asks for "the full arsenal", "use everything you have", "/Apeiron", "/delegate-team", "delegate to a backend", "spawn a multi-agent team", "make a CLI to scaffold skills", or any natural-language request that should drive a Claude Code or Codex session through the full brainstorm → plan → execute → verify → ship pipeline. Composes four companion frameworks (superpowers, Waza, unslop-preflight, autoresearch) plus the original `dt` gateway CLI, multi-agent team (MMAS) framework, skill scaffolder, the /Apeiron orchestrator (47-case selftest routing), and the v2.7.0 arsenal: autopilot.sh (7-stage GOD command), apeiron-uni (smart universal wrapper that detects the calling runtime), and 1890 auto-discovered skills across 3 sources.

2026-07-171

agent-kernel

imMamdouhaboammar/delegate-team

Local-first governance kernel for AI coding agents — gives Claude Code, Codex, Cursor, Antigravity, and Gemini CLI a shared memory, an approval inbox for new rules, generated AGENTS.md / CLAUDE.md / cursor-rule.mdc, Claude + git hooks, and a deterministic policy guard. Adds a JSON-first memory layer (rules + preferences + workflows + project-notes + skills), an episodic memory archive, and MCP tools for search/read/capture/sync of past sessions. Use when the user asks to "remember this rule", "save this as a memory", "what did we do last time", "search past episodes", "propose a new rule", "make agent follow our standards automatically", or any request to give an agent persistent local memory across sessions. Bundled inside delegate-team v2.5.0 — installed via `./install.sh --kernel`. Triggers on: "agent kernel", "ak", "remember this", "save this rule", "memory tool", "episodic memory", "approval inbox", "guard policy", "rule inbox".

2026-07-171

aonios-agent

imMamdouhaboammar/delegate-team

Route heavy coding tasks to premium multi-model backends (Codex GPT-5.5/5.4 high, GLM-5.2 max, Qwen 3.7 max, Kimi K2.7 code-max, MiniMax M3 high-thinking). Mirrors vertex-coder's structure but shells out to local CLI binaries (codex, opencode) instead of Vertex AI. Two modes: direct (single-file) and interactive (multi-file autonomous with tool use). Use when delegating heavy code generation, debugging, or refactoring tasks to the strongest available model. Triggers on: "delegate to aonios agent", "use the strongest model", "premium model coding", "opencode glm", "codex gpt-5 high", "minimax m3 thinking", "aonios agent direct", "aonios agent interactive". Do NOT use for trivial edits (do them locally) or when the user explicitly forbids external delegation.

2026-07-171

minimax-coder

imMamdouhaboammar/delegate-team

Local MiniMax-powered coding agent. Uses the official `mmx` CLI (https://github.com/MiniMax-AI/cli) to call MiniMax models (M3 flagship, M2.7, M2.7-highspeed) for direct single-file edits and multi-file autonomous work with full tool use. Sibling backend to vertex-coder inside delegate-team; mirrors its structure (direct + interactive + tools registry) but routes through mmx instead of an SDK. The Python scripts in this directory are local-agent helpers — Apeiron (M3) can also invoke `mmx text chat` directly via bash for fast ad-hoc calls. Three MiniMax models supported. Use when delegating coding work to MiniMax models. Triggers on: "use minimax coder", "MiniMax M3", "MiniMax M2.7", "minimax direct", "minimax interactive", "mmx text chat". Do NOT use for trivial edits (do them locally).

2026-07-171

apeiron-team

imMamdouhaboammar/delegate-team

Apeiron Multi-Agent System (MMAS) — spawn a team of specialized AI agents under your direct command. Each agent has a specific power + model (Atlas orchestrator, Forge deep worker, Scout searcher, Oracle debugger, Librarian docs-finder, Reviewer validator, Visionary media-analyzer, Sentinel security-auditor). A watchdog polls every 30s and reports status back to you. Inspired by oh-my-openagent's specialized agent team + hermes-agent's self-improving learning loop. Use when you want parallel agents working on a complex task while you stay the boss. Triggers on: "spawn team", "multi-agent", "MMAS", "parallel agents", "team of agents", "boss mode", "watchdog", "/apeiron-team". Do NOT use for single-file edits (do them locally).

2026-07-171

apeiron

imMamdouhaboammar/delegate-team

2026-07-171

name	apeiron
description	The ONE-COMMAND orchestrator that wires the full delegate-team arsenal for any task in any project. Use when the user says "use everything you got", "use full engineering arsenal", "go all-in", "/Apeiron <task>", "/ship <task>", "use all capabilities", "full stack", "end to end", "everything intelligently". Composes 38 curated integrations (superpowers, Waza, unslop, autoresearch, agent-kernel, ...) and routes tasks through 7-stage autopilot chain (PREWARM → BRAINSTORM → PLAN → EXECUTE → REVIEW → QUALITY-GUARD → REPORT). Auto-discovers 1890+ skills across 3 sources (apeiron, agents, claude) with intelligent dedupe. Includes apeiron-uni — the smart universal wrapper that detects the calling runtime (Apeiron / codex / claude / gemini / opencode / mmx / shell) and dispatches the right flow. Skip this skill only when the user wants a SPECIFIC tool (e.g. "run /autoresearch with metric X" or "delegate to aonios-agent").
allowed-tools	["Bash","Read","Write","Edit","Grep","Glob","TodoWrite","Task","Skill","WebFetch","WebSearch"]

/Apeiron — the orchestrator

The single switch that exposes the full delegate-team arsenal under one natural-language command. Type /Apeiron "<task>" and the orchestrator does the rest.

What this skill IS

A bundle that exposes the full Apeiron arsenal under one natural-language command. Bundles:

orchestrate.py — 47-case routing selftest; routes by task signature to the right chain (FEATURE / BUG / UI / PERFORMANCE / RESEARCH / TRIVIAL / MULTI-AGENT).
catalog.py — 38 curated integrations + 1890 unique skills auto-discovered from ~/.apeiron/skills/, ~/.agents/skills/, ~/.claude/skills/ (with dedupe).
autopilot.sh — the GOD command: 7-stage chain (PREWARM → BRAINSTORM with superpowers + codex gpt-5.5-high → PLAN → EXECUTE → REVIEW → QUALITY-GUARD → REPORT). Runs in foreground or --background.
apeiron-uni — smart universal wrapper: detects calling runtime and dispatches the right flow (Apeiron → prewarm; agents → autopilot --background; shell → autopilot foreground).

Outcome contract

Outcome: A task entered as natural language → routed through the right stack → shipped (or audit-blocked if quality gates fail).
Done when: every stage green OR explicit user override, with evidence at each step.
Evidence: skill outputs, unslop audit score, autoresearch deltas, code review findings, quality-guard pass.
Output: shipped code (or blocked-task report with reasons).

When to use this skill (the DEFAULT)

User said	Orchestrator does
`<any task>` and they want full leverage	full chain, route by signature
"use everything", "all-in", "go crazy"	full chain
"I have a feature / bug / perf task"	abbreviated chain based on signature
`<UI task>`	unslop audit is BLOCKING (score ≥70)
`<trivial edit>`	skip chain, handle locally

When NOT to use this skill

User said	Use instead
"run /autoresearch with metric X"	`/autoresearch` directly
"delegate to aonios-agent"	`/delegate-team aonios-agent` directly
"spawn atlas + forge"	`/apeiron-team atlas forge` directly
"explain X to me"	direct answer, no orchestration
"research X"	`/read` + `/learn` only

Usage

From any Apeiron session (canonical):

Apeiron "<task>"
# → orchestrate emits JSON manifest → Apeiron loads skills → executes

From any agent (codex, claude, gemini, opencode, mmx):

apeiron-uni "<task>"
# → autopilot --background (returns PID + log path; agent can move on)

From plain shell:

apeiron-uni "<task>"
# → autopilot foreground (you watch the log scroll)

Direct (bypass the smart wrapper):

# Get the route + plan only (no execution)
orchestrate.py --prewarm "<task>"

# Run the full chain in foreground
autopilot.sh "<task>"

# Run the full chain in background
autopilot.sh "<task>" --background

# Dry-run (just show the plan)
autopilot.sh --dry-run "<task>"

The 7-layer chain (autopilot.sh)

[Apeiron "<task>"]
    │
    ▼ 1. PREWARM (orchestrate.py + catalog.py)
       Detect task signature → load relevant skills → emit manifest JSON
       Backend selection: codex (PERFORMANCE), claude (UI), minimax (BUILD/BUG/FEATURE)
    │
    ▼ 2. BRAINSTORM (superpowers:brainstorming + codex gpt-5.5-high)
       The mix: superpowers enforces 'no diving into code without a design'.
       Codex gpt-5.5-high provides the highest-quality ideation.
    │
    ▼ 3. PLAN (superpowers:writing-plans)
       Convert brainstorm into checkpoint-style plan with evidence gates.
    │
    ▼ 4. EXECUTE (delegate-team → backend agent)
       Hand off to dt with the brief. Backend writes the code.
    │
    ▼ 5. REVIEW (waza:check)
       Project-aware review with evidence. Surface constraints.
    │
    ▼ 6. QUALITY-GUARD (5-layer pre-delivery check)
       Layer 1: mechanical (lint/typecheck/test/build)
       Layer 2: definition-of-done
       Layer 3: security
       Layer 4: AI-smells
       Layer 5: project-specific
    │
    ▼ 7. REPORT
       Markdown report with all evidence + next steps for the user.

Routing intelligence (47 cases, 100% selftest passing)

signature = detect(task)
# detect():
#   has_metric = re.search(r'(\d+)%|p\d+|p \d+|< \d|>\d+|reduce.*by|increase.*by', task)
#   is_ui = 'frontend' in task or 'ui' in task or 'design' in task or 'page' in task
#   is_bug = 'fix' in task or 'broken' in task or 'regression' in task
#   is_research = 'research' in task or 'learn' in task or 'understand' in task
#   is_trivial = len(task) < 60 and 'fix' not in task

if is_research:
    return ["read", "learn"]
if is_trivial:
    return ["local"]
if is_bug:
    return ["think", "systematic-debugging", "hunt", "delegate-team", "check"]
if has_metric:
    return ["think", "autoresearch:plan", "autoresearch:fix", "autoresearch:regression"]
if is_ui:
    return ["think", "unslop", "writing-plans", "delegate-team", "check", "quality-guard"]
return ["think", "writing-plans", "delegate-team", "check", "quality-guard"]

Worked examples

# 1. Performance (PERFORMANCE path)
Apeiron "Make API p95 < 200ms"
# → autoresearch loop is the engine, then quality-guard

# 2. UI feature (UI DELIVERY path)
Apeiron "Build a beautiful settings page with dark mode"
# → unslop audit is BLOCKING (score≥70) before delegate-team

# 3. Bug fix (BUG path)
Apeiron "Mobile header wrong on Safari iOS 17"
# → debug-issue before any patch, then quality-guard

# 4. Plain feature (FEATURE path)
Apeiron "Build a TypeScript CLI that converts CSV to JSON"
# → delegate-team (→ mini-coder-max), then quality-guard

# 5. Trivial (TRIVIAL path)
Apeiron "rename getCurrentUser to getActiveUser across src/"
# → handle locally, skip the chain

Files in this skill

SKILL.md — this file (skill manifest)
scripts/orchestrate.py — router (47-case selftest, v4.1.0)
scripts/orchestrate.sh — bash wrapper that delegates to orchestrate.py
scripts/catalog.py — 38 integrations + 1890 skills indexer
scripts/.DS_Store, scripts/__pycache__/ — local artifacts (gitignored)

Helper scripts

The apeiron-uni + autopilot.sh scripts are also part of this arsenal but live in ~/delegate-team/bin/ (the central symlink bin). They are installed alongside this skill by install.sh (the delegate-team installer).

Companion frameworks (lazy-loaded on demand)

Framework	What it adds	Install via
superpowers	Methodology — brainstorm-first hard gate, TDD, review	bundled
Waza	Entry-point skills — /think /check /hunt /ui	bundled
unslop-preflight	UI quality gate — 23 gates block generic slop	bundled
autoresearch	Metric-driven iteration loop	bundled
agent-kernel	Local-first memory + governance layer	bundled
delegate-team	The CLI gateway + multi-model router	bundled

Compatibility

Agent	Install	Status
Apeiron session	`install.sh --all`	✅ canonical
Claude Code	`npx skills add imMamdouhaboammar/delegate-team`	✅ via `~/.claude/skills/apeiron/`
Codex	`npx skills add ...`	✅ via `~/.codex/skills/`
Cursor	`npx skills add ...`	✅ via `~/.agents/skills/`
Copilot, Windsurf, OpenCode, 60+ more	`npx skills add ...`	✅ via `~/.{tool}/skills/`

License

MIT — see https://github.com/imMamdouhaboammar/delegate-team

Repository

https://github.com/imMamdouhaboammar/delegate-team/tree/master/apeiron