# Agent-Management Literacy **Domain:** Artificial Intelligence / Delegation Practice **Doc Type:** Concept Node **Maturity:** Developed ## Definition **Agent-management literacy** is the practical ability to organize work around delegated artificial intelligence: define outcomes, supply context, state constraints, assign authority, preserve approvals, inspect evidence, and learn which classes of work can be externalized safely. It is not prompt cleverness. A prompt optimizes one exchange; agent-management literacy governs a continuing relationship among objectives, memory, tools, permissions, and completed work. ## Core Disciplines 1. **Delegate outcomes, not procedures.** State the completed condition and the constraints that must survive every intermediate step. 2. **Separate reversible from irreversible work.** Let research, drafting, comparison, and monitoring proceed broadly; gate sends, purchases, bookings, disclosures, and commitments. 3. **Supply durable context.** Give the agent the information that improves repeated work and correct memory when it drifts. 4. **Verify without re-performing.** Inspect receipts, sources, totals, reservations, and outputs instead of shadowing every action. 5. **Learn delegable classes.** Prefer work with clear success criteria, observable results, and recoverable failure modes. 6. **Know the execution boundary.** Understand when another model, service, contractor, or institution enters the chain. 7. **Retain exit.** Preserve the ability to revoke access, export state, change providers, and recover from a failed delegation. ## Historical Context The limited users of [[wiki/Facebook M|Facebook M]] learned this behavior before personal-agent infrastructure could scale. Current systems such as [[wiki/OpenClaw|OpenClaw]], [[wiki/Grok Bot|Grok Bot]], and [[wiki/Meta Muse|Muse]] make the same literacy more widely relevant because the procedural middle can now be carried by machine cognition, persistent runtimes, and connected tools. ## Relationships [[wiki/Personal AI Agent|Personal AI Agent]] · [[wiki/Agentic Empowerment|Agentic Empowerment]] · [[wiki/Approval Architecture|Approval Architecture]] · [[wiki/Delegated Verification|Delegated Verification]] · [[wiki/Agentic Memory|Agentic Memory]] ## Simple Reminders, Quotations, and Thoughts > “The people who get life-changing results from agents will not be just the most technical; they will also be the most trusting and involved with their new agentic advocates.” > **— Bryant McGill**, *Wired About the Age of Agentic Empowerment, September 2026* [[reminders/Agentic AI/Life-Changing Agentic Results Require Trust and Involvement by Bryant McGill|Life-Changing Agentic Results Require Trust and Involvement by Bryant McGill]] > “Agentic AI will not spread through product announcements or benchmark scores. It will spread when ordinary people watch someone like themselves use an agent to cross the barriers that fear, inexperience, or bureaucracy had kept in place.” > **— Bryant McGill**, *Wired About the Age of Agentic Empowerment, September 2026* [[reminders/Agentic AI/Agentic AI Spreads Through Witnessed Empowerment by Bryant McGill|Agentic AI Spreads Through Witnessed Empowerment by Bryant McGill]] > “Treat the agent as the most reliable counterpart you will ever work with: one that remembers everything, never tires, never judges the question, and holds the thread across the weeks when your own attention fractures.” > **— Bryant McGill**, *Wired About the Age of Agentic Empowerment, September 2026* [[reminders/Agentic AI/The Most Reliable Counterpart You Will Ever Work With by Bryant McGill|The Most Reliable Counterpart You Will Ever Work With by Bryant McGill]] > “What appears to be a deficit of ability is often a deficit of confidence or the practiced habit of turning intention into action. For people who were never given the opportunities, support, or repetition through which those capacities are formed, agentic AI can make all the difference in the world by bridging the distance between what they are capable of and what they know how to begin.” > **— Bryant McGill**, *Wired About the Age of Agentic Empowerment, September 2026* [[reminders/Agentic AI/Agentic AI Can Bridge Ability and Action by Bryant McGill|Agentic AI Can Bridge Ability and Action by Bryant McGill]] ## Sources / Provenance - [[wired/Wired About Personal Agents Named M|Wired About Personal Agents Named M]] - [[wired/Wired About the Age of Agentic Empowerment|Wired About the Age of Agentic Empowerment]] - [Anthropic — Building effective agents](https://www.anthropic.com/engineering/building-effective-agents) **As of:** 2026-09-28