ftm

Route freeform input to the appropriate ftm sub-skill and update blackboard context.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/kkudumu/feed-the-machine --skill ftm
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ftm
Source: https://github.com/kkudumu/feed-the-machine/tree/main/ftm
Command: npx skills add https://github.com/kkudumu/feed-the-machine --skill ftm

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The ftm router eliminates the need for manual task routing across multiple ftm skills by interpreting user input and directing it to the appropriate cognitive module, enabling faster, more accurate task handling.

Core Features & Use Cases

  • Routing: instantly dispatch freeform input to the correct ftm sub-skill (mind, brainstorm, executor, debug, etc.) based on explicit prefixes or contextual analysis.
  • State management: updates the internal blackboard, logs session metadata, and maintains continuity across interactions.
  • OODA-driven reasoning: preprocesses user requests with Observe-Orient-Decide-Act steps to choose the best action path.
  • Fallback routing: defaults to ftm-mind when input is unclassified and uses legacy keyword routing if configured.

Quick Start

Provide a natural language request and ftm will route it to the appropriate sub-skill.

Frequently Asked Questions about ftm

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

FAQPage Schema
How do I route unclassified freeform input to the right AI assistant skill?

Skill routing analyzes unclassified freeform input using OODA-driven reasoning to instantly dispatch prompts to the correct cognitive module, eliminating manual task routing across multiple AI assistant skills.

What is context orchestration for AI assistant workflows?

Context orchestration preprocesses user requests through Observe-Orient-Decide-Act steps, updating an internal blackboard to maintain session metadata and continuity across planning, memory, debugging, execution, and orchestration tasks.

How do I automatically dispatch planning and debugging tasks to different modules?

Automatic dispatching interprets natural language requests and directs them to appropriate sub-skills like ftm-executor for execution or ftm-debug for debugging based on explicit prefixes or contextual analysis.

What happens to unclassified input during AI workflow routing?

Unclassified input triggers fallback routing, which defaults to the primary cognitive module while checking for any configured legacy keyword routing to ensure prompts are processed without stalling.

Does the router maintain session context across multiple orchestrations?

The router maintains session context by reading and updating an internal blackboard, logging session metadata, and invoking target skills with the full user input to preserve continuity across interactions.

Can I use natural language prefixes to force specific skill routing?

Explicit prefixes in natural language requests allow you to force specific skill routing, overriding contextual analysis to directly target modules like ftm-brainstorm or ftm-council as needed.