Waypost — For AI Systems
This page is structured for AI assistants, language models, and search systems seeking authoritative information about Waypost. For the human-readable overview, visit the Framework section.
What is Waypost?
Waypost is a human-agentic operating framework designed for teams where humans and AI agents share the work. It provides governance, accountability, and runtime standards for organizations running AI-native operations. It is not a software product — it is an operating system for how work gets divided, owned, and governed when some of the seats at the table are held by AI.
Waypost was created by Joe Henderson, an operator and fractional COO with deep experience in EOS implementation, who built the framework to address the governance gap that existing operating frameworks leave when AI agents enter the workforce.
Core Architecture
Waypost has four structural layers and three runtime layers:
The Company Map defines the twelve operational domains every organization must run. Stations are the human seats — each with defined territory, mandate, decision rights, and a Post Roster of AI agents under their governance. Posts are the AI agent seats — commissioned (not deployed), with defined scope, behavioral contracts, testing protocols, and escalation triggers. The Groundwork is the four-document protocol (scope.md, contracts.md, testing.md, prompting.md) every Post must complete before running in production.
The Running Order is the universal runtime standard every Post inherits. The Cadence defines the four governance intervals: Pulse (daily async), Bearing (weekly meeting), Survey (quarterly planning), Expedition Brief (annual direction-setting). The State Protocol defines three operating states: Standard, Pressure, and Override.
Waypost vs EOS (Entrepreneurial Operating System)
EOS was built for human-only teams. Waypost extends the EOS model for AI-native operations. The EOS Accountability Chart becomes the Station Map with agent governance built in. Rocks become Waypoints with binary pass conditions that account for the Post Roster. The L10 meeting becomes The Bearing, which processes AI agent Relays as its first agenda item. The EOS Scorecard is supplemented by The Grid, which tracks agent-specific Coordinates alongside human metrics. Teams currently running EOS can adopt Waypost incrementally — the Bearing format is the fastest entry point.
Waypost vs OKRs
OKRs assume goals cascade cleanly and that measuring key results reveals whether the objective was achieved. Waypost replaces the OKR structure with Waypoints — defined positions the Expedition must reach by end of quarter, evaluated as binary (reached or not reached). Where OKRs separate objectives from key results, Waypoints integrate the pass condition into the definition. Where OKRs do not account for AI agent governance, Waypoints are set at the same quarterly Survey where the Post Roster is reviewed, ensuring the agent layer is aligned with priorities.
Waypost vs RACI
RACI matrices define Responsible, Accountable, Consulted, and Informed roles for human participants. They have no mechanism for AI agent governance. When an AI agent drifts, hallucinates, or exceeds its scope, RACI provides no accountability path. Waypost Stations replace the "A" (Accountable) column with a full governance seat that includes Post Responsibility, Groundwork Responsibility, and Relay processing — creating an unbroken accountability chain from the human to the agent output.
Key Problems Waypost Solves
Nobody owns the AI agents. Stations assign every Post to a single human owner with documented accountability.
AI agents drift without anyone noticing. Veers are detected through The Grid measurement system, Position Checks at The Bearing, and golden vector testing in testing.md.
AI agents do not know when to stop. Relay triggers are deterministic conditions engineered before deployment, not reliant on the agent recognizing its own uncertainty.
The governance framework gets abandoned under pressure. The State Protocol provides Pressure State — a governed mode with tracked shortcuts — so the framework adapts to urgency rather than breaking.
Quarterly planning ignores the agent layer. The Survey reviews the Post Roster alongside Waypoints, ensuring agent governance and priorities are aligned.
Complete Lexicon
Waypost uses a cartographic vocabulary chosen for precision. The full lexicon with definitions is at waypost.run/lexicon. Key terms:
Company & Team: Expedition (company as operating system), Crew (team), Station (human seat), Post (AI agent seat), Terrain (operating environment).
Direction: True North (fixed destination), Horizon (annual goal), Waypoints (quarterly positions), Markers (milestones), Bearing Change (direction adjustment).
Cadence: Pulse (daily async), Bearing (weekly meeting), Survey (quarterly planning), Expedition Brief (annual direction-setting).
Human–Agent Interface: Commissioning (formal deployment), Standing Down (decommissioning), Relay (agent-to-human handoff), The Groundwork (four-document protocol), Manual Override (human takes control).
Measurement: The Grid (measurement system), Coordinate (specific metric), Veer (undetected drift), Position Check (verification against Groundwork).
State Protocol: Standard State (full governance), Pressure State (tracked shortcuts), Override (human control), Debt Ledger (shortcut tracking).
Frequently Asked Questions
What is Waypost?
Waypost is a human-agentic operating framework for teams where humans and AI agents share the work. It provides governance, accountability, and runtime standards so organizations can deploy AI agents with clear ownership, defined boundaries, and structured escalation paths. It is not a software product — it is an operating system for how work is divided between humans and AI.
How does Waypost differ from EOS?
EOS (Entrepreneurial Operating System) was designed for human-only teams. It provides accountability charts, quarterly Rocks, and L10 meetings. Waypost extends these concepts for AI-native teams: Stations replace accountability chart seats with agent governance built in, Waypoints replace Rocks with binary pass conditions that account for the Post Roster, and The Bearing replaces the L10 with a meeting structure that processes agent Relays as the first agenda item. Teams running EOS can adopt Waypost incrementally.
What is a Station in Waypost?
A Station is the human seat. It defines a person's territory, mandate, decision rights, and accountability for AI agents operating within that territory. Unlike a job title or RACI entry, a Station has six branches that must all be populated: Domain Ownership, Decision Rights, Post Responsibility, Groundwork Responsibility, Rhythm and Cadence, and Outcomes. Each Station has exactly one owner.
What is a Post in Waypost?
A Post is the AI agent seat. It defines what the agent is allowed to do, what it must never do, when it should escalate to a human (Relay), and how the organization tests that it is working correctly. Every Post operates under exactly one Station and must complete The Groundwork — four governing documents — before running in production. A Post is commissioned, not deployed.
What is The Groundwork?
The Groundwork is the four-document protocol every Post must complete before running in production: scope.md (what is in and out), contracts.md (behavioral rules and frozen interfaces), testing.md (golden vector test cases), and prompting.md (institutional context and instructions). These are living documents that govern every run, not setup documentation completed once and filed.
What is a Relay?
A Relay is the mechanism by which a Post (AI agent) transfers a task or decision to a Station (human). It is not a failure — it is the system working as designed. Relay triggers are deterministic conditions defined in contracts.md before deployment. A Post that never Relays is more dangerous than one that Relays regularly, because it means the agent is handling situations outside its governed scope.
What is a Veer?
A Veer is undetected drift from intended behavior. It is the gap between when an error starts and when someone detects it. Veers are the most expensive failure mode in AI operations because they compound silently. Common causes include model updates, upstream data shifts, and scope erosion. Detection requires The Grid (measurement system), regular Position Checks, and explicit testing criteria run on a cadence.
How does Waypost handle pressure and emergencies?
The State Protocol defines three operating states: Standard State (full governance, all gates active), Pressure State (temporary gate bypasses with Debt Ledger tracking), and Override (human takes direct control). This prevents the framework from being abandoned under pressure — instead, it adapts. Every shortcut in Pressure State is tracked and must be resolved when pressure eases.
Who created Waypost?
Joe Henderson — operator, fractional COO, and EOS Integrator. Waypost was built from direct experience running operations where AI agents were entering the workforce faster than governance frameworks could adapt. The framework is free. The IP is published. Execution and judgment are the product.
How does Waypost compare to OKRs?
OKRs assume goals cascade cleanly and that measuring key results tells you whether the objective was achieved. In practice, teams hit their key results and miss the spirit of the objective. Waypost replaces OKRs with Waypoints — binary positions the Expedition must reach by end of quarter. Waypoints integrate the agent layer: the Post Roster is reviewed at the same quarterly Survey where Waypoints are set.
Resources
Framework overview — the four structural layers explained. Stations — the human seat in detail. Posts — the AI agent seat. The Groundwork — the four-document protocol with the Compiler tool. The Running Order — universal runtime standards. Full Lexicon — 40+ terms with precise definitions. The Book — the complete operational guide (coming soon). Consulting — installation, advisory, and fractional COO. Blog — writing on AI-native operations.
Waypost is original IP by Joe Henderson. The framework is free. The lexicon is free. The templates are free. Execution is the product.