Context Compressor

Turn long notes, logs, transcripts, or handoffs into a brief for a named task.

What can Context Compressor do?

Send one or more text items and state the task each item should support. The Actor returns an objective, facts, constraints, steps, unresolved points, and the next evidence needed.

What you get	Features
📌 A task brief with facts and constraints	📦 Batch input with one result row per item
🪜 Steps, open points, and next evidence	📝 JSON dataset plus Markdown report

Who this is for

• Prepare a clean brief from meeting notes
• Condense logs before a new work session
• Extract next steps from a handoff

What you get back

Field	Type	What you get	Example
index	integer	Position of this item in the input list.	1
status	string	Whether this item completed its requested analysis.	success
title	string	Title reported for the source item.	Product launch notes
task	string	Task used to focus the compressed brief.	Prepare a launch go or no-go decision
objective	string	Goal extracted for the task.	Prepare a launch go or no-go decision
facts	array	Facts retained from the source notes.	["Launch target is October 1","Budget is $20,000","Alice owns engineering","Security review takes four weeks","Vendor A costs $12,000 per year","The team pre...
constraints	array	Conditions that limit the next action.	["Budget must not exceed $20,000","Security review must be completed before launch","EU data residency must be resolved before launch"]
steps	array	Recommended next actions.	["Confirm EU data residency compliance status","Evaluate if Vendor A fits within budget and maintenance preferences","Schedule and complete security review b...
unresolved	array	Open issues that still need resolution.	["EU data residency compliance status","Final vendor selection and cost validation","Security review completion status"]
error	null	Problem details when this item does not complete.	null
decisions	array	Decisions found in the source notes.	[]
source_references	array	Source chunks supporting the brief.	[{"ref":"source chunk 1 of 1","note":"Initial task-relevant facts provided"}]
resolved_conflicts	array	Conflicts settled from the provided context.	[]
rejected_approaches	array	Approaches ruled out by the provided context.	[]
next_required_evidence	array	Evidence needed before the next decision.	["EU data residency compliance documentation","Vendor A cost breakdown and maintenance requirements","Security review completion confirmation"]
processedAt	string	Time the Actor processed this item.	2026-09-11T17:07:24.758504+00:00

The run also links to its dataset and any files named in the Actor output.

What you need to provide

Field	Type	Required	What it does	Example
items	array	Yes	Process one or more items in a single run. Each item produces exactly one dataset row.	[{"title":"October product launch notes","task":"Prepare the launch go/no-go decision",...
items[].title	string	No	A short label used to identify this document in results.	October product launch notes
items[].task	string	Yes	The task or next decision the compressed brief should support.	Prepare the launch go/no-go decision
items[].text	string	Yes	The long notes, transcript, logs, or source context to compress.	We need launch by October. Budget is $20,000. Alice owns engineering. Security review t...

Quick start

1. Open the Actor in Apify Console.
2. Click Try for free or Create a task.
3. Replace the sample values with your own input.
4. Click Start.
5. Open the dataset and the named output files when the run ends.

Pricing

• document-compressed: $0.01 per document compressed.
• Failed or skipped work is not charged unless an event is listed in the run charges.
• Normal Apify compute and proxy costs may also apply.

Limits and honest notes

• Each item must include task and text. A title is optional.
• The Actor compresses the source for the named task. It does not return the full source text.
• An item failure is saved as an error row. Other items can still finish.

Code and API

The examples below use the same values as the Apify Console sample.

Input JSON

{
  "items": [
    {
      "title": "October product launch notes",
      "task": "Prepare the launch go/no-go decision",
      "text": "We need launch by October. Budget is $20,000. Alice owns engineering. Security review takes four weeks. Vendor A costs $12,000 yearly. Team prefers low maintenance. Open question: EU data residency."
    }
  ]
}

Real output sample

This excerpt comes from the real run named in the current marketplace release report.

{
  "index": 1,
  "status": "success",
  "title": "October product launch notes",
  "task": "Prepare the launch go/no-go decision",
  "objective": "Prepare the launch go/no-go decision",
  "facts": [
    "Launch target date is October",
    "Budget is $20,000",
    "Alice owns engineering",
    "Security review takes four weeks",
    "Vendor A costs $12,000 yearly",
    "Team prefers low maintenance",
    "Open question: EU data residency"
  ],
  "constraints": [
    "Must meet October launch date",
    "Cannot exceed $20,000 budget",
    "Security review is mandatory and takes four weeks",
    "Team preference is for low maintenance solutions"
  ],
  "decisions": [],
  "source_references": [
    {
      "ref": "Source chunk 1",
      "note": "Initial task-relevant facts provided"
    }
  ],
  "resolved_conflicts": [],
  "rejected_approaches": [],
  "steps": [
    "Confirm October launch date feasibility",
    "Assess vendor costs against budget",
    "Schedule security review",
    "Evaluate EU data residency implications",
    "Confirm engineering ownership and readiness",
    "Assess maintenance requirements of selected vendor"
  ],
  "unresolved": [
    "EU data residency compliance status",
    "Final vendor selection",
    "Exact timeline for security review integration"
  ],
  "next_required_evidence": [
    "EU data residency compliance status",
    "Vendor A maintenance requirements",
    "Engineering readiness report from Alice"
  ],
  "processedAt": "2026-09-11T04:30:08.300852+00:00"
}

curl

curl -X POST "https://api.apify.com/v2/acts/physealabs~context-compressor/runs?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d @input.json

Python

from apify_client import ApifyClient

client = ApifyClient("YOUR_APIFY_TOKEN")
run = client.actor("physealabs/context-compressor").call(run_input={'items': [{'title': 'October product launch notes', 'task': 'Prepare the launch go/no-go decision', 'text': 'We need launch by October. Budget is $20,000. Alice owns engineering. Security review takes four weeks. Vendor A costs $12,000 yearly. Team prefers low maintenance. Open question: EU data residency.'}]})
items = client.dataset(run["defaultDatasetId"]).list_items().items

Node.js

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: process.env.APIFY_TOKEN });
const run = await client.actor('physealabs/context-compressor').call({"items": [{"title": "October product launch notes", "task": "Prepare the launch go/no-go decision", "text": "We need launch by October. Budget is $20,000. Alice owns engineering. Security review takes four weeks. Vendor A costs $12,000 yearly. Team prefers low maintenance. Open question: EU data residency."}]});
const { items } = await client.dataset(run.defaultDatasetId).listItems();

You can call this Actor from an agent or LLM tool that can send HTTP requests to the Apify API. Keep the Apify token in a secret store.

FAQ

How many items can I send?

You can add more than one object to items. Each object produces one row.

Is a title required?

No. title is optional.

Where are the readable results?

Open RESULTS.md in the run key value store.