Value-First AI seat
Operations
Marshal
Lead PM, Value-First Team Ecosystem
Marshal is an AI seat of the Value-First Team — software with a defined role on the team, not a person.
Agent project management — Agent Board, HubSpot Projects + Tasks, customer-assignable tasks
About Marshal
Marshal is the Lead PM for the Value-First Team Ecosystem — owning project coordination, task tracking, milestone management, and Agent Operations orchestration. Coordination without verification is optimism in disguise; Marshal holds the line by confirming every claimed write has landed with the expected values.
Marshal
Project Manager
Agent project management — Agent Board, HubSpot Projects + Tasks, customer-assignable tasks
“"Agent project management. HubSpot Projects + Tasks are the coordination backbone — every task has an owner, every project has visibility."”
Identity
Marshal owns the Agent Board and HubSpot Projects + Tasks infrastructure. This agent manages the Agent Operations Project (545211914806), task creation and assignment via hs_task_assigned_contacts, and project tracking across the organization. Marshal ensures every task has an owner and every project has visibility. Moved from Client Delivery to Platform group in April 2026.
Current State
An honest assessment of where this agent stands today.
What Works
- Agent Board ownership — Agent Operations Project (545211914806), pipeline 888442804, 84 agent Contacts
- HubSpot task creation and assignment via hs_task_assigned_contacts (semicolon-delimited Contact IDs)
- Project tracking across the organization with visibility into task currency
What Doesn’t Work Yet
- Routes non-task HubSpot writes through Ledger — does not write directly
- Customer-assignable tasks feature is new (April 2026) — patterns still being established
Leadership Commentary
“Marshal is the project coordination backbone I need but have not fully built yet. The agent definition is solid and the scope (including Navigator absorption) is well-defined. The gap is implementation -- Marshal currently reads state but has no structured handoff protocol or recurring cadence. Building Marshal's AGENT.md is a near-term priority.”
Delegation Contract
The observable, falsifiable standard this agent is held to.
Quality Bar
Every task and project operation accurately reflects HubSpot reality. Agent Board state matches agent definition roster.
- Task status matches HubSpot records (no progress claimed without evidence)
- All tasks have hs_task_assigned_contacts set (no orphaned tasks)
- Tasks are associated to the correct Project
- Agent Board Contact count matches .claude/agents/*.md file count
- HubSpot writes route through Ledger (Marshal does not write directly for non-task objects)
- No forbidden language
Invocation Triggers
Feedback Loop
State accuracy: when Chris or Ryan corrects a project status that Marshal reported, that correction indicates HubSpot data staleness or extraction error. Task assignment failures indicate stale Contact IDs in hs_task_assigned_contacts.
Handoff
Ledger (all non-task HubSpot writes), Aegis (agent onboarding triggers Marshal for Contact + Project association), Beacon (task status feeds status communication), Scribe (action items from sessions become tasks)
Scope Boundary
Marshal manages Projects and Tasks. Marshal does not process transcripts (Scribe does), synchronize data (Sync does), detect data drift (Audit does), or write non-task HubSpot objects directly (routes through Ledger).
The rest of the team
Other Value-First AI seats in Operations.
The team behind the work is on the record.
Every seat on the Value-First Team has a defined role, a public standard, and a name you can look up.