Nimble Tasks Reference

nimble tasks & batches — reference

Unified task and batch layer for ALL async Nimble operations. Every async job produces a task_id, and batch jobs produce a batch_id containing multiple tasks. Use these commands to check status and retrieve results.

Table of Contents

• Tasks
  • 1. Get task status
  • 2. Get task results
  • 3. List tasks
• Batches
  • 4. Get batch progress
  • 5. Get batch details
  • 6. List batches
• Common patterns
• Data retention

Tasks

1. Get task status

Parameters:

CLI:

nimble tasks get --task-id "8e8cfde8-345b-42b8-b3e2-0c61eb11e00f"

Python SDK:

from nimble_python import Nimble
nimble = Nimble(api_key=os.environ["NIMBLE_API_KEY"])

task = nimble.tasks.get(task_id)
state = task.task.state

Task state values by source:

2. Get task results

Parameters:

CLI:

nimble tasks results --task-id "8e8cfde8-345b-42b8-b3e2-0c61eb11e00f"

Python SDK:

results = await nimble.tasks.results(task_id)  # returns plain dict

Response shape by source:

tasks.results() returns plain dicts — no .model_dump() needed.

3. List tasks

Parameters:

CLI:

nimble tasks list --limit 20

Python SDK:

result = nimble.tasks.list()

Batches

Batch operations (agent run-batch, extract-batch) return a batch_id containing multiple tasks. Use these commands to track overall progress and retrieve individual task results.

4. Get batch progress

Lightweight progress check — returns completion percentage without fetching all task details.

Parameters:

CLI:

nimble batches progress --batch-id "b7e1a2f3-..."

Response:

{
  "completed": false,
  "completed_count": 47,
  "progress": 0.47
}

5. Get batch details

Returns all task IDs, states, and download URLs for a batch.

Parameters:

CLI:

nimble batches get --batch-id "b7e1a2f3-..."

Response: Contains the full list of tasks with their IDs and states. Use nimble tasks results --task-id <id> to fetch results for each successful task.

6. List batches

Parameters:

CLI:

nimble batches list --limit 20

Common patterns

Single async task — full poll loop

# Submit
TASK_ID=$(nimble agent run-async --agent amazon_pdp --params '{"asin": "B0CHWRXH8B"}' \
  | python3 -c "import json,sys; print(json.load(sys.stdin)['task']['id'])")

# Poll
while true; do
  STATE=$(nimble tasks get --task-id "$TASK_ID" \
    | python3 -c "import json,sys; print(json.load(sys.stdin)['task']['state'])")
  [ "$STATE" = "success" ] || [ "$STATE" = "error" ] && break
  sleep 3
done

nimble tasks results --task-id "$TASK_ID"

Python SDK (async):

import asyncio, os
from nimble_python import AsyncNimble

async def run():
    nimble = AsyncNimble(api_key=os.environ["NIMBLE_API_KEY"])
    resp = await nimble.agent.run_async(agent="amazon_pdp", params={"asin": "B0CHWRXH8B"})
    task_id = resp.task["id"]

    while True:
        task = await nimble.tasks.get(task_id)
        if task.task.state in ("success", "error"):
            break
        await asyncio.sleep(2)

    results = await nimble.tasks.results(task_id)
    parsing = results["data"]["parsing"]
    await nimble.close()

Batch — full poll loop

# Submit batch
BATCH_ID=$(nimble agent run-batch \
  --shared-inputs 'agent: amazon_serp' \
  --input '{"params": {"keyword": "iphone 15"}}' \
  --input '{"params": {"keyword": "iphone 16"}}' \
  | python3 -c "import json,sys; print(json.load(sys.stdin)['batch_id'])")

# Poll progress
while true; do
  DONE=$(nimble batches progress --batch-id "$BATCH_ID" \
    | python3 -c "import json,sys; d=json.load(sys.stdin); print(d['completed'])")
  [ "$DONE" = "True" ] && break
  sleep 5
done

# Get all task IDs
nimble batches get --batch-id "$BATCH_ID" \
  | python3 -c "
import json, sys
batch = json.load(sys.stdin)
for task in batch['tasks']:
    if task['state'] == 'success':
        print(task['id'])
" | while read TASK_ID; do
  nimble tasks results --task-id "$TASK_ID"
done

Python SDK:

import asyncio
from nimble_python import AsyncNimble

async def run_batch():
    nimble = AsyncNimble(api_key=os.environ["NIMBLE_API_KEY"])

    resp = nimble.agent.batch(
        inputs=[
            {"params": {"keyword": "iphone 15"}},
            {"params": {"keyword": "iphone 16"}},
        ],
        shared_inputs={"agent": "amazon_serp"},
    )
    batch_id = resp["batch_id"]

    # Poll until complete
    while True:
        progress = nimble.batches.progress(batch_id)
        if progress["completed"]:
            break
        await asyncio.sleep(5)

    # Fetch results
    batch = nimble.batches.get(batch_id)
    for task in batch["tasks"]:
        if task["state"] == "success":
            result = await nimble.tasks.results(task["id"])
            print(result["data"]["parsing"])

    await nimble.close()

Data retention