Execute an API Workflow with osop run

OSOP Guide

Run an API SOP workflow end-to-end. Real HTTP requests with headers, body, auth. Cost controls. Security gates. Execution log generated.

6 nodes · 7 edgesosop self

osopthe-loopapiexecutemeta

Visual

Select .osop Workflowhuman

Choose which API SOP to execute. Example: stripe-payment-flow.osop.yaml

↓sequential→ Pre-flight Risk Assessment

Pre-flight Risk Assessmentagent

AI scans the workflow for security risks: destructive commands, missing approval gates, broad permissions, hardcoded secrets. Returns a risk score 0-100.

↓conditional→ Approve Execution

↓conditional→ osop run

Approve Executionhuman

Review the risk assessment. Approve or reject. CLI nodes require --allow-exec flag.

↓sequential→ osop run

osop runcli

Execute the workflow. API nodes send HTTP requests with headers, body, query params. Agent nodes call LLMs. Results flow between nodes via WorkflowContext.

↓sequential→ .osoplog Generated

↓fallback→ Select .osop Workflow

.osoplog Generatedsystem

Execution record: per-node duration, cost, status, HTTP status codes, response previews, AI token usage, human decisions.

↓sequential→ osop report

osop reportcli

Generate a standalone HTML report from the execution log. Open in any browser.

Open in Editor

ex-osop-run-api-workflow.osop.yaml

osop_version: "1.0"
id: "osop-run-api-workflow"
name: "Execute an API Workflow with osop run"
description: "Run an API SOP workflow end-to-end. Real HTTP requests with headers, body, auth. Cost controls. Security gates. Execution log generated."
tags: [osop, the-loop, api, execute, meta]

nodes:
  - id: "select_workflow"
    type: "human"
    name: "Select .osop Workflow"
    description: "Choose which API SOP to execute. Example: stripe-payment-flow.osop.yaml"

  - id: "preflight_check"
    type: "agent"
    subtype: "llm"
    name: "Pre-flight Risk Assessment"
    description: "AI scans the workflow for security risks: destructive commands, missing approval gates, broad permissions, hardcoded secrets. Returns a risk score 0-100."

  - id: "human_approve"
    type: "human"
    subtype: "review"
    name: "Approve Execution"
    description: "Review the risk assessment. Approve or reject. CLI nodes require --allow-exec flag."
    security:
      approval_gate: true
      risk_level: "medium"

  - id: "execute"
    type: "cli"
    name: "osop run"
    description: "Execute the workflow. API nodes send HTTP requests with headers, body, query params. Agent nodes call LLMs. Results flow between nodes via WorkflowContext."
    runtime:
      command: "osop run workflow.osop.yaml --allow-exec --max-cost 1.00 --log"

  - id: "log_generated"
    type: "system"
    name: ".osoplog Generated"
    description: "Execution record: per-node duration, cost, status, HTTP status codes, response previews, AI token usage, human decisions."

  - id: "view_report"
    type: "cli"
    name: "osop report"
    description: "Generate a standalone HTML report from the execution log. Open in any browser."
    runtime:
      command: "osop report workflow.osop.yaml execution.osoplog.yaml -o report.html"

edges:
  - from: "select_workflow"
    to: "preflight_check"
    mode: "sequential"
  - from: "preflight_check"
    to: "human_approve"
    mode: "conditional"
    condition: "preflight_check.risk_score > 30"
  - from: "preflight_check"
    to: "execute"
    mode: "conditional"
    condition: "preflight_check.risk_score <= 30"
  - from: "human_approve"
    to: "execute"
    mode: "sequential"
  - from: "execute"
    to: "log_generated"
    mode: "sequential"
  - from: "log_generated"
    to: "view_report"
    mode: "sequential"
  - from: "execute"
    to: "select_workflow"
    mode: "fallback"
    label: "Execution failed, retry"

←

Import Existing Workflows into SOP Doc

SOP Doc: Browse and Understand Any Workflow

→