openma Agent Creator
What is openma?
openma is an open-source platform for building, deploying, and managing AI agents.
Think of it as a managed runtime — you define an agent (model + system prompt + tools),
the platform handles sandboxed execution, credential management, and session state.
What you can do with it:
- Build agents for any task: coding, research, data analysis, customer support, automation
- Run agents in sandboxes — each session gets an isolated container with file system, shell, and network
- Connect external services via MCP servers (GitHub, Slack, Linear, Notion, etc.) with OAuth
- Use any LLM — Anthropic, OpenAI, DeepSeek, or any OpenAI-compatible provider
- Install community skills from ClawHub to extend agent capabilities
- Manage credentials securely in vaults — agents get scoped access, secrets never leak
- Collaborate — multi-user workspace with API key access for CLI/SDK integration
Creating an Agent
Flow
Understand the goal — ask what the agent should do. If vague, one question:
"What's the main task?" Two rounds max, then build.
Pick the model — check /v1/model_cards first. Defaults:
- Complex/coding:
claude-opus-4-6
- General (default):
claude-sonnet-4-6
- Simple/fast:
claude-haiku-4-5-20251001
- OpenAI:
gpt-4o, o3
Write system prompt — specific, actionable, bounded. Not generic.
Select tools — default agent_toolset_20260401 (file ops, bash, web) covers most cases.
Create:
POST /v1/agents
{ "name", "model", "system", "tools": [{"type":"agent_toolset_20260401"}] }
Next steps — offer to create session, configure skills, set up model card.
Platform Quick Ref
Agents need a session to run. Sessions need an environment (sandbox).
| Resource |
What it is |
| Agent |
Model + system prompt + tools config |
| Session |
A conversation with an agent in a sandbox |
| Environment |
Sandbox runtime (default works for most) |
| Model Card |
API key + provider config for an LLM |
| Vault |
Secure credential storage for MCP/CLI secrets |
| Skill |
SKILL.md that gives agents domain expertise |
| API Key |
Programmatic access token for CLI/SDK |
oma agents create <name> # create agent
oma sessions create --agent <id> --env <id> # start session
oma sessions message <id> <text> # send message
oma models create --name <n> --model-id <id> --api-key <key>
oma keys create # generate API key
oma skills install <slug> # install from ClawHub
oma --help # full command list
Model card providers: ant, oai, ant-compatible, oai-compatible.
Source: open-ma/open-managed-agents — distributed by TomeVault.
1---2name: open-ma-open-managed-agents-open-managed-agents3description: openma Agent Creator4---56# openma Agent Creator78## What is openma?910openma is an open-source platform for building, deploying, and managing AI agents.11Think of it as a managed runtime — you define an agent (model + system prompt + tools),12the platform handles sandboxed execution, credential management, and session state.1314**What you can do with it:**15- **Build agents** for any task: coding, research, data analysis, customer support, automation16- **Run agents in sandboxes** — each session gets an isolated container with file system, shell, and network17- **Connect external services** via MCP servers (GitHub, Slack, Linear, Notion, etc.) with OAuth18- **Use any LLM** — Anthropic, OpenAI, DeepSeek, or any OpenAI-compatible provider19- **Install community skills** from ClawHub to extend agent capabilities20- **Manage credentials** securely in vaults — agents get scoped access, secrets never leak21- **Collaborate** — multi-user workspace with API key access for CLI/SDK integration2223## Creating an Agent2425### Flow26271. **Understand the goal** — ask what the agent should do. If vague, one question:28 "What's the main task?" Two rounds max, then build.29302. **Pick the model** — check `/v1/model_cards` first. Defaults:31 - Complex/coding: `claude-opus-4-6`32 - General (default): `claude-sonnet-4-6`33 - Simple/fast: `claude-haiku-4-5-20251001`34 - OpenAI: `gpt-4o`, `o3`35363. **Write system prompt** — specific, actionable, bounded. Not generic.37384. **Select tools** — default `agent_toolset_20260401` (file ops, bash, web) covers most cases.39405. **Create:**41 ```42 POST /v1/agents43 { "name", "model", "system", "tools": [{"type":"agent_toolset_20260401"}] }44 ```45466. **Next steps** — offer to create session, configure skills, set up model card.4748## Platform Quick Ref4950Agents need a **session** to run. Sessions need an **environment** (sandbox).5152| Resource | What it is |53|----------|-----------|54| Agent | Model + system prompt + tools config |55| Session | A conversation with an agent in a sandbox |56| Environment | Sandbox runtime (default works for most) |57| Model Card | API key + provider config for an LLM |58| Vault | Secure credential storage for MCP/CLI secrets |59| Skill | SKILL.md that gives agents domain expertise |60| API Key | Programmatic access token for CLI/SDK |6162```63oma agents create <name> # create agent64oma sessions create --agent <id> --env <id> # start session65oma sessions message <id> <text> # send message66oma models create --name <n> --model-id <id> --api-key <key>67oma keys create # generate API key68oma skills install <slug> # install from ClawHub69oma --help # full command list70```7172Model card providers: `ant`, `oai`, `ant-compatible`, `oai-compatible`.7374---75> Source: [open-ma/open-managed-agents](https://github.com/open-ma/open-managed-agents) — distributed by [TomeVault](https://tomevault.io).76<!-- tomevault:4.0:skill_md:2026-06-26 -->
tomevault-io