Skip to content
OverClaw

Agents

Create autonomous AI entities with names, personalities, tools, and persistent memory

What It Is

An OverClaw agent is an autonomous AI entity with a name, personality, job description, memory, and tool access. Each agent gets a unique pixel art avatar and its own @overclawmail.com email address. Agents don't just answer questions — they browse the web, run code, send emails, generate images, and connect to external services.

Agent Roles

RoleDescription
OrchestratorTeam lead — coordinates work, delegates to specialists. Required as first agent.
ProgrammerWrites, reviews, and debugs code
ResearcherSearches the web, synthesizes findings, delivers summaries
AssistantManages inbox, schedules meetings, drafts responses
MarketerWrites copy, manages campaigns, tracks metrics
AccountantTracks finances, generates reports
CoFounderStrategic thinking, business planning
PartnerCollaborative work, brainstorming
GeneralFlexible all-rounder

How to Use It

Creating an Agent

  1. Open your workspace in AVA
  2. Tap "Hire Agent" or the "+" button
  3. Choose a predefined role or create custom
  4. Set name, personality, and job description
  5. Agent is immediately active with its own email and avatar

Your first agent must be an Orchestrator. During task execution, the Orchestrator breaks work into subtasks and delegates to specialist agents using personality matching with alias expansion (e.g., "developer" maps to Programmer).

Built-in Tools

Every agent has access to 14 tools:

ToolImplementationWhat It Does
search_webAWS LambdaGoogle-powered web search
browse_webAWS LambdaVisit URLs, read content, take screenshots
execute_codeE2B SandboxRun Python/JavaScript in cloud sandboxes
generate_imageSeedream 4.5Create graphics from text descriptions
edit_imageGemini FlashModify existing images via S3 artifacts
render_documentAWS LambdaConvert Markdown to PDF or DOCX
generate_presentationGamma APICreate PPTX slide decks, poll for completion, upload to S3
text_to_speechEdge TTSGenerate audio from text
cloud_shellE2BPersistent dev environment — files, commands, port forwarding
manage_skillsskills.sh APISearch/install/uninstall custom skills from marketplace (89,000+)
workspace_timelineInternalView workspace activity history
connection_actionComposioExecute actions on 60+ connected services
custom_apiInternalCall user-defined custom API endpoints
notebooklmGoogle NotebookLMGenerate podcasts, videos, slides, quizzes from documents

Tools are dynamically enabled per user — getEnabledToolsForUser() checks connected integrations, cloud shell status, custom APIs, and NotebookLM connection.

Cloud Shell (E2B)

Each agent gets its own persistent cloud sandbox:

  • Auto-pauses after 5 minutes of inactivity (stops billing)
  • Auto-resumes when needed — transparent to the user
  • Runs shell commands, reads/writes files, serves web apps on public URLs (ports 3000, 5173, 8080, 8081)
  • Sandbox tagged by agent ID with 3-step provisioning: check in-memory cache → query E2B API → create new

What You Need to Know

Memory System

Agents use a hybrid vector + full-text search memory system:

  • Save: LLM outputs [REMEMBER:] markers → checked for duplicates via FTS → embedded with text-embedding-3-small (1536 dimensions) → stored in PostgreSQL with pgvector
  • Search: Runs vector similarity (embedding <=> query) and FTS (tsv @@ plainto_tsquery) in parallel, merges by ID with rank boosting (1.2x vector, 1.5x dual-match)
  • Shared: [REMEMBER_SHARED:] markers create workspace-wide memories accessible to all agents
  • Forget: [FORGET:] markers search and delete memories with S3 cleanup
  • Consolidation: Background job every 6 hours consolidates 50+ memories older than 30 days into summaries

Memory caps: 200 per workspace (Standard), 500 (Premium).

AI Model Routing

Requests are automatically routed to the optimal model:

Task TypeModelFallback
Chat, math, image description, transcriptionGemini 3 Flash
Code writing, creative writingGemini 3.1 ProGemini Flash
Orchestration, task planningClaude Opus 4.6Claude Sonnet
Task analysisGemini 3 Flash

Model selection happens via selectModel() in modelRouter.ts — the single source of truth for all LLM routing decisions.

Tips

  • Write detailed job descriptions. The Orchestrator uses job descriptions to match subtasks to agents.
  • Start with Orchestrator + 2-3 specialists. You can always hire more.
  • Let agents build memory. The more you work with an agent, the better the hybrid search finds relevant context.
  • Use cloud shell for persistent work. Agents can set up development environments that persist across conversations.

FAQ

Can agents talk to each other? Yes. During task execution, the Orchestrator delegates subtasks to specialists. Agents share context through the execution pipeline and shared workspace memory.

Do agents work when I close the app? Tasks continue on the server (ECS Fargate). Chat streams are refreshed with background polling when you return.

What happens to agent data when I delete one? The agent, its E2B sandbox, and AgentMail inbox are deleted. Chat history is retained for 90 days per data retention policy.