Invocation
Session start, context recovery, mission briefing, and dispatch. Invocation covers the critical transition from cold start to productive collaboration: how context is restored, objectives are communicated, and agents are activated for specific missions.
8 methods| ID | Method | Description |
|---|---|---|
| FG-0401 | Context Recovery Protocol 4 | Define a structured sequence for restoring agent context at the start of every session. AI agents start each session with no memory of previous work. Context Recovery Protocol establishes a systematic loading sequence that brings agents up to speed efficiently: read governance first, then project state, then current priorities, then specific task context. The order matters because each layer informs the interpretation of the next. |
| FG-0402 | Parallel Agent Dispatch | Invoke multiple specialized agents simultaneously for complex tasks that span multiple domains. Parallel Agent Dispatch is multi-agent orchestration applied to a single mission: identify the domains involved, select the appropriate specialists, brief them simultaneously, and coordinate their parallel execution. This maximizes throughput by using the model's ability to work on multiple fronts at once. |
| FG-0403 | Session Rituals | Establish consistent start-of-session and end-of-session practices that create rhythm and ensure completeness. Session Rituals are not ceremony for ceremony's sake: they serve operational functions. A standup at session start ensures alignment on priorities. A closeout at session end ensures work is committed, documented, and ready for the next session. Consistency creates reliability. |
| FG-0404 | Mission Briefing Format | Structure task assignments using a standardized briefing format that ensures completeness. A Mission Briefing is a structured task assignment that includes: objective (what), context (why), constraints (boundaries), resources (what's available), success criteria (how you'll know it's done), and authority level (what decisions the agent can make autonomously). This format eliminates ambiguity and reduces the need for clarifying questions mid-task. |
| FG-0405 | Custom Command System | Build a library of reusable, parameterized commands that trigger complex agent behaviors with simple invocations. Custom Commands encode frequently-used multi-step operations into single triggers. Instead of explaining a complex procedure every session, the team invokes a named command that expands into the full procedure. This reduces errors, saves time, and creates a shared operational vocabulary. |
| FG-0406 | Invocation Authority | Define who can invoke which agents and for what purposes. Not every participant should be able to dispatch every agent, especially in multi-human teams. Invocation Authority establishes which roles can activate which agents, preventing unauthorized use of specialized or sensitive capabilities. This is the access control layer for agent dispatch. |
| FG-0407 | Multi-Session Orchestration 3 | Running multiple agent sessions simultaneously, each focused on a different task or domain, coordinated through shared state and artifact tracking. The pattern is parallel dispatch: independent work streams that produce artifacts consumed by other streams, with a human operator routing decisions between sessions. It mirrors operational environments where multiple teams execute simultaneously under a single commander's direction. |
| FG-0408 | Dual-Lane Session Architecture 3 | Running parallel session lanes with distinct domain boundaries, where each lane has its own identity, scope, and update protocol. The key insight beyond basic multi-session is that sessions need domain boundaries as well as coordination protocols. Each lane owns a domain and cannot accidentally modify the other's state. Coordination happens through shared state files rather than direct session interaction. |