NEXUS Agent Utilities — Apify Actor

Deterministic utilities for AI agents, automation pipelines, CI/QA workflows, and machine-to-machine data processing.

The Actor runs without external paid AI APIs. Each operation is predictable, lightweight, and produces structured output in the default dataset.

Operations

• normalize_json — recursively sorts object keys and normalizes strings for stable machine-to-machine comparisons.
• clean_lines — trims text, removes blank lines, and deduplicates lines while preserving order.
• json_diff — produces a path-based structural diff between two JSON objects.
• text_qa — checks duplicate lines, trailing whitespace, long lines, and control characters.
• code_risk_scan — performs a defensive static pattern scan for code-review signals. It never executes supplied code.

Quick examples

Normalize JSON

{
  "operation": "normalize_json",
  "data": {
    "z": " value ",
    "a": 1
  }
}

Clean lines

{
  "operation": "clean_lines",
  "text": "alpha\n\n beta \nalpha\ngamma"
}

Compare JSON

{
  "operation": "json_diff",
  "left": {"status": "draft", "count": 1},
  "right": {"status": "ready", "count": 2}
}

Text QA

{
  "operation": "text_qa",
  "text": "first line\nsecond line  \nsecond line"
}

Defensive code-risk scan

{
  "operation": "code_risk_scan",
  "code": "const input = process.env.USER_INPUT;"
}

Good use cases

• AI-agent preprocessing and post-processing
• deterministic QA before an LLM call
• stable JSON normalization for caching, hashing, and regression tests
• text cleanup before ingestion or embeddings
• JSON change detection in automation pipelines
• lightweight defensive checks in CI/review workflows

Pricing

NEXUS Agent Utilities uses Apify Pay Per Event.

• Utility call: $0.20 per 1,000 calls
• Actor start: $0.00005
• Apify platform usage is included in the event pricing shown in the Store

The primary billable event is utility-call, emitted only for completed deterministic utility operations.

Output

Each successful operation writes one structured result to the default dataset. The output includes the selected operation, success state, and deterministic result payload.

Design principles

• no external paid AI APIs
• no execution of user-supplied code
• deterministic behavior
• small, composable operations
• suitable for API, workflow, and agent integration
• limited Actor permissions

API and automation

Use the Actor from Apify Console, the Apify API, tasks, schedules, or other Actors. The input schema exposes only the fields required by the selected operation.

Support

If an operation produces an unexpected result, open an issue from the Actor page and include the operation name plus a minimal reproducible input. Do not include secrets or credentials.