agentgrid-sessions
Skill9 starsChanged 5 months agoMITView on GitHub
Save, restore, and manage agentgrid sessions and presets — persist grid layouts, replay company configurations, and use built-in presets. Use when "save grid", "restore session", "load preset", "replay company" is needed.
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 107
Don't rely on sessions for persistent state — agent context is not saved, only the grid structure.
- Distinctive: line 108
Don't restore a session into a tmux window that already has a grid — kill existing panes first.
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
When to Use
- Save a working grid layout to replay it later (or overnight)
- Restore a previous company configuration after a crash or new terminal
- Load a bundled preset for common team patterns (dev-sprint, research-swarm, full-company)
- List what sessions/presets are available
Session Commands
Save a grid
# Save current grid layout, pane labels, and agent types
agentgrid save my-company
# Sessions are stored in ~/.agentgrid/sessions/Restore a grid
# Restore a previously saved session
agentgrid restore my-company
# This re-creates panes, re-applies labels, and re-launches agentsList saved sessions
agentgrid preset list # lists both built-in presets and saved sessionsBuilt-in Presets
Presets live in presets/ in the agentgrid package. Launch any with:
agentgrid preset load <name>| Preset | Panes | Description |
dev-sprint | 4 | CEO + 2 builders + QA |
full-company | 6 | CEO + VP + 3 builders + QA |
research-sprint | 3 | CEO + 2 researchers |
research-swarm | 5 | CEO + 4 parallel researchers |
anti-drift-squad | 3 | CEO + auditor + fixer |
design-sprint | 4 | CEO + designer + builder + QA |
earning-factory | 4 | CEO + 3 earning agents |
mixed-agents | 4 | Mixed: Claude + Codex + Gemini + aider |
sparc-pipeline | 5 | SPARC: spec + pseudocode + arch + refine + complete |
Typical Session Workflow
# 1. Create and configure a grid
agentgrid 3x2 "claude --dangerously-skip-permissions --chrome"
# 2. Name each pane (CEO runs in each pane)
agentgrid name "CEO"
# 3. Do your work...
# 4. Save for later
agentgrid save energy-fleet
# — Next session —
# 5. Restore
agentgrid restore energy-fleet
# Agents are re-launched in their labeled panes automaticallyOvernight Autonomous Pattern
# Save after grid is configured and working
agentgrid save overnight-run
# If the session crashes or context degrades, restore:
agentgrid restore overnight-run
# Auto-switch.sh integrates with sessions:
# ./scripts/auto-switch.sh --overnight
# It reads .claude/handoff.md and calls agentgrid restore if a session was savedSession Storage Location
Sessions are stored as JSON in ~/.agentgrid/sessions/. Each file contains:
- Grid dimensions
- Pane labels (names)
- Agent command per pane
- Working directory per pane
You can inspect or version-control these files directly.
Anti-Patterns
- Don't rely on sessions for persistent state — agent context is not saved, only the grid structure.
- Don't restore a session into a tmux window that already has a grid — kill existing panes first.
- Presets launch fresh agents; combine with
agentgrid injectto re-assign missions.
In this repository31 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- llms.txtllms.txt
- agentgrid-orchestrateSkill
- agentgridSkill
- agent-runtimeSkill
- frontend-designSkill
- architectSkill
- deep-thinkSkill
- design-systemsSkill
- ecosystem-referenceSkill
- fractal-delegationSkill
- harness-reviewSkill
- heartbeatSkill
- integrate-resourcesSkill
- llm-council-critiqueSkill
- loop-createSkill
- mcp-setupSkill
- memory-compressionSkill
- memorySkill
- model-routingSkill
- oss-project-scaffoldSkill
- playwright-cliSkill
- repo-scoutSkill
- self-improveSkill
- skill-creatorSkill
- skill-routingSkill
- sprintSkill
- test-visualSkill
- troubleshootSkill
- ux-auditSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
Reviews can't be read right now.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.