opcode
Zero-token execution layer for AI agents. Define workflows once, run them free forever — persistent, scheduled, deterministic. 6 MCP tools over SSE. Supports DAG-based execution, 6 step types (action, condition, loop, parallel, wait, reasoning), 26 built-in actions, ${{}} interpolation, reasoning nodes for human-in-the-loop decisions, and secret vault. Use when defining workflows, running templates, checking status, sending signals, querying workflow history, or visualizing DAGs.
Why use this skill?
Automate recurring AI tasks with OPCODE. Define persistent, deterministic DAG workflows to eliminate repeat token burn and streamline complex agent agent orchestration.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/rendis/opcodeWhat This Skill Does
OPCODE is a zero-token execution layer designed to transform OpenClaw agents from reactive assistants into proactive, deterministic automated systems. By defining workflows once, users can offload repetitive, multi-step logic—such as data processing pipelines, recurring status reports, or complex CI/CD orchestration—directly to the OPCODE daemon. Once a workflow is defined as a JSON-based Directed Acyclic Graph (DAG), the agent no longer needs to "reason" through the steps for every run; it simply triggers the pre-defined sequence. This eliminates unnecessary token burn and ensures identical, deterministic output across every execution instance. The engine supports 6 distinct step types, including actions, conditions, parallel processing, and human-in-the-loop reasoning nodes.
Installation
To integrate OPCODE with your environment, first install the binary via Go: go install github.com/rendis/opcode/cmd/opcode@latest. Perform the first-time setup with opcode install --listen-addr :4100 --vault-key "your-passphrase". Ensure your OpenClaw MCP configuration includes the server pointing to http://localhost:4100/sse. You can also install the skill directly via the Clawhub repository using clawhub install openclaw/skills/skills/rendis/opcode.
Use Cases
OPCODE is ideal for persistent background tasks. Use it for automated financial report generation where data must be pulled from multiple APIs, sanitized, and formatted weekly. It is also highly effective for multi-agent coordination, such as a content-writer agent drafting a document and a deploy-bot agent managing the publication lifecycle. Because workflows persist after process restarts, it is perfect for long-running monitoring tasks that require state awareness.
Example Prompts
- "Define a new workflow template named 'weekly-report' that first fetches analytics data, waits for a reasoning node approval, and then emails the summary to the team."
- "List all suspended workflows currently managed by the opcode daemon and summarize why they were halted."
- "Execute the 'deploy-pipeline' workflow with the parameters set to production mode and notify me once the deployment completes."
Tips & Limitations
Always choose a stable agent_id for your workflows, as these IDs are used for provenance in the history logs. Remember that while OPCODE reduces inference tokens, it requires an active daemon process to function; always secure your vault-key to prevent unauthorized execution of your defined workflows. For debugging, use the opcode.diagram tool frequently to visualize the DAG structure of complex workflows before triggering them in production environments.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-rendis-opcode": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: network-access, file-write, file-read, code-execution