skills.rest
berrzebb
berrzebbCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

gateway-direct-execution

Classify gateway requests and route them through direct execution paths.

22|7|Updated Feb 23, 2026
Search Tags:#routing#execution#gateway#direct execution#request classification#result envelope
One-click install
npx skills add https://github.com/berrzebb/SoulFlow-Orchestrator --skill gateway-direct-execution
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: gateway-direct-execution
Source: https://github.com/berrzebb/SoulFlow-Orchestrator/tree/main/src/skills/gateway-direct-execution
Command: npx skills add https://github.com/berrzebb/SoulFlow-Orchestrator --skill gateway-direct-execution

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill streamlines the process of handling incoming requests by intelligently routing them through the most efficient and secure execution paths, minimizing reliance on complex agentic workflows when simpler methods suffice.

Core Features & Use Cases

  • Intelligent Request Routing: Classifies requests into categories like builtin, direct_tool, model_direct, workflow_compile, workflow_run, and agent_required.
  • Direct Execution Optimization: Prioritizes direct execution paths for tools, models, or workflows when an agent is not strictly necessary, reducing latency and cost.
  • Standardized Output: Ensures all results, regardless of the execution path, are formatted into a standard ResultEnvelope.
  • Channel Affinity: Guarantees that responses are returned to the exact channel the user initiated the request from.
  • Use Case: When a user asks to execute a predefined tool, this skill ensures it's run directly without invoking a full agent, and the result is returned in a clean format to the user's chat.

Quick Start

Use the gateway-direct-execution skill to classify an incoming request and determine the most efficient execution path.

Frequently Asked Questions about gateway-direct-execution

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I route gateway requests directly to tools without invoking a full agent?▼

Gateway request routing classifies incoming requests into predefined types like direct_tool or model_direct, executing them immediately without an agent. This direct execution optimization reduces latency and minimizes computational cost for simple tasks.

What is the best way to standardize outputs from different gateway execution paths?▼

Standardizing gateway outputs uses a ResultEnvelope to format all execution responses uniformly. Regardless of whether a request triggers a direct tool, model interaction, or workflow run, the result is packaged into this standard structure.

How does request classification determine if a workflow needs to compile or run?▼

Request classification evaluates incoming gateway requests and categorizes them as workflow_compile or workflow_run based on the required action. This ensures the gateway routes the request through the correct execution path for workflow processing.

Can I maintain channel affinity for responses when using direct execution paths?▼

Direct execution paths maintain channel affinity by guaranteeing that all responses are returned to the exact channel the user initiated the request from. This ensures the standardized ResultEnvelope reaches the correct destination.

When should I avoid direct execution and use an agent_required routing path instead?▼

You should use the agent_required routing path when a request is too complex for direct tool execution, model-direct interactions, or workflow compilation. The gateway classifies such requests to ensure they are handled by a full agentic workflow.

Related Skills

Explore More Software Engineering Skills→
swaruplabswaruplab
94

flowio

Parse FCS files and export events to NumPy arrays or DataFrames.

Community
Advanced
photonics-dhlphotonics-dhl

group-chat-output-standard

Format group-chat results into standardized report-ready output with literature links.

Community
Intermediate
parcadeiparcadei
3.9k

implement_plan

Execute technical plans with risk assessment and multi-phase verification.

Community
Advanced
lushly-devlushly-dev
4

afd-directclient

Execute commands between co-located AI agents within a Node.js process.

Official
Advanced