Agent skill
fleet-deployment
Use when deploying agents from YAML configuration files
Install this agent skill to your Project
npx add-skill https://github.com/nouamanecodes/lettactl/tree/main/.skills/fleet-deployment
SKILL.md
Entry Points
src/commands/apply.ts- Deploy fleet from YAMLsrc/commands/validate.ts- Validate config syntaxsrc/lib/fleet-parser.ts- YAML parsing & content resolutionsrc/lib/diff-engine.ts- Change detectionsrc/lib/diff-applier.ts- Apply changes to Letta
Commands
lettactl apply -f <file> [--dry-run] [--force] [-q] [-v]
lettactl apply-template -f <file> --template <pattern> [--dry-run]
lettactl validate -f <file>
Key Types
FleetConfig {
root_path?: string
shared_blocks?: SharedBlock[]
mcp_servers?: McpServerConfig[]
agents: AgentConfig[]
}
AgentConfig {
name: string
description: string
system_prompt: { value?: string; from_file?: string; from_bucket?: BucketConfig }
llm_config: { model: string; context_window: number }
tools?: string[]
shared_blocks?: string[]
memory_blocks?: MemoryBlock[]
folders?: FolderConfig[]
}
MemoryBlock {
name: string
description: string
limit: number
value?: string
from_file?: string
agent_owned?: boolean // default: true
}
Examples
# Deploy fleet
lettactl apply -f fleet.yaml
# Preview changes
lettactl apply -f fleet.yaml --dry-run
# Remove resources not in config
lettactl apply -f fleet.yaml --force
# Apply template to existing agents
lettactl apply-template -f template.yaml --template "^prod-.*"
Environment
LETTA_BASE_URL- Letta server URL (required)LETTA_API_KEY- API key (optional for self-hosted)SUPABASE_URL,SUPABASE_ANON_KEY- For cloud storage
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
lettactl
Manage Letta AI agent fleets with kubectl-style CLI
release-cycle
Use when releasing features - covers issues, branches, tests, commits, PRs, versioning, and publishing
resource-management
Use when managing resources like memory blocks, tools, folders, files, or MCP servers
message-operations
Use when sending messages to agents, viewing conversation history, or managing message state
code-review
Thorough code review covering architecture, code quality, and performance. Use when asked to "review this plan", "review code", "audit architecture", "check code quality", or "review for performance". Walks through issues interactively with tradeoff analysis and opinionated recommendations.
bulk-operations
Use when performing bulk operations like cleanup, delete-all, or bulk messaging
Didn't find tool you were looking for?