ftm-mind

Orchestrate an Observe-Orient-Decide-Act loop with memory and context.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/kkudumu/feed-the-machine --skill ftm-mind
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ftm-mind
Source: https://github.com/kkudumu/feed-the-machine/tree/main/ftm-mind
Command: npx skills add https://github.com/kkudumu/feed-the-machine --skill ftm-mind

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

ftm-mind provides a unified Observe-Orient-Decide-Act loop to reason, orient around live state, and plan before acting, enabling memory across conversations and safer, more deliberate task execution.

Core Features & Use Cases

  • Memory-aware reasoning: remembers session context, experiences, and patterns to inform decisions.
  • Fast orientation: loads current state, prerequisites, and constraints before acting.
  • End-to-end task handling: supports freeform /ftm or explicit skill invocations, Jira-driven work, and multi-tool workflows.
  • Safe execution: promotes plan-first/approval gating when needed and updates blackboard after tasks.

Quick Start

Provide a freeform request such as '/ftm-mind help me think through this issue' to start the full OODA loop.

Frequently Asked Questions about ftm-mind

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

FAQPage Schema
How does an OODA loop help with planning before acting on vague tasks?

Memory-aware reasoning retains session context, experiences, and patterns to inform decisions across conversations. It loads current state and prerequisites before acting, enabling safer and more deliberate task execution with continuity.

How do I start a planning workflow with a freeform request?

Start a planning workflow by providing a freeform request such as '/ftm-mind help me think through this issue'. The orchestrator initiates the full OODA loop, loading context and routing to appropriate sub-skills while honoring prerequisites and approval gates.

Can I use this orchestrator for Jira-driven work and mixed-tool workflows?

Yes, the orchestrator requires a YAML frontmatter containing the name and description fields. This metadata is necessary to define the task parameters and route correctly to the appropriate ftm sub-skills during the orientation phase.

What are the limits of plan-first approval gating in task execution?

The limits of plan-first approval gating involve the overhead of loading context and prerequisites before acting. While it promotes safer execution for vague tasks, it may introduce delays for straightforward tasks that do not require upfront understanding or multi-step orchestration.