Supported Log Export File Formats

Filestack log exports are delivered in NDJSON (newline-delimited JSON), with each line representing a single event record from the processing pipeline. This format was chosen because it streams cleanly, parses line-by-line without loading entire archives into memory, and integrates directly with common SIEM and log-analytics platforms used by compliance teams. Alongside the core NDJSON export, gzip-compressed variants are available for high-volume accounts, reducing storage overhead for retained logs.

Also worth reading: How do I export Filestack audit logs via API? · What is the BYOK cost pricing model for Filestack? · How does Filestack ensure GDPR compliant data retention for enterprise compliance teams?

Beyond NDJSON, exports can be rendered as CSV for teams that need spreadsheet-friendly outputs for audit reviews, though nested fields are flattened into dot-notation columns in that case. All formats carry identical event data, including timestamps, user identifiers, transformation actions, and integrity hashes, so the choice of format affects only downstream handling rather than content. Exports are generated per retention window and signed with a checksum, letting compliance teams verify that exported logs match the source records byte-for-byte before archiving them.

Exporting Compliance Audit Logs

The unresolved part of Filestack's log export story is the format itself, and it's now settled: exports are delivered as newline-delimited JSON (JSONL), one event per line, encoded as UTF-8. Each record carries a timestamp in ISO 8601, an event type, the actor identity, the resource handle, and a request correlation ID, so downstream SIEM tools can ingest the file without transformation. The same schema is used whether you trigger an export from the dashboard or pull it through the API, which keeps tooling consistent across workflows.

For compliance teams, JSONL was the pragmatic choice because it streams cleanly, tolerates partial failures during large exports, and is accepted natively by most log aggregation platforms. If your pipeline requires CSV or syslog instead, you can convert after ingestion, but the canonical export remains JSONL. Retention windows and export frequency are configured per workspace, and every export itself is logged as an auditable event, closing the loop on chain-of-custody requirements.

Retention and Storage Options

When logs are exported from the platform, they are written as newline-delimited JSON (NDJSON), with each event serialized as a single JSON object on its own line. This format keeps every field intact — timestamps, actor identifiers, action types, and resource references — so compliance teams can parse exports with standard tooling or pipe them directly into a SIEM without transformation. Exports can be delivered to a connected S3 bucket or retrieved on demand through the dashboard.

For teams with longer retention requirements, the underlying storage layer supports archival rules that move older export files to cold storage automatically, while the logs themselves remain immutable once written. That immutability matters in audit contexts, since it allows you to demonstrate that records were not altered after the fact. If you need a specific schema sample or a signed URL example from an actual export, let me know and I can pull one from the documentation or a sandbox environment.

Pipeline Event Field Definitions

The export format Filestack's log pipeline produces is newline-delimited JSON, commonly called NDJSON. Each pipeline event is written as a single compact JSON object on its own line, with no pretty-printing or indentation, which keeps the file streamable and easy to ingest into downstream tools like Splunk, Elasticsearch, or a data lake. Every record carries a consistent field set: a timestamp in ISO 8601 format with UTC offset, an event type string, the request or file identifier, the storage region, and the actor context such as API key hash or user session. Optional fields, like transformation metadata or policy results, appear only when relevant, so consumers should treat absent keys as null rather than expecting them on every line.

Files are delivered gzip-compressed with a .ndjson extension and rolled on a hourly boundary, so a full day yields twenty-four archive segments. Schema versions are declared through a top-level "schema_version" integer, letting compliance teams pin parsers to a known contract before upgrading.

Validating Exports for Audits

One open question about Filestack's logging has been whether audit exports arrive in a structured machine-readable format or as plain text dumps. For compliance teams that must feed pipeline activity into SIEM tools, this matters. Filestack's activity logs export as JSON by default, with each event captured as a discrete object containing timestamps, event types, resource identifiers, and actor metadata. That structure means the exported log can be ingested directly by common audit tooling without intermediate parsing, and fields can be mapped to your retention schema on arrival.

For teams preferring CSV for spreadsheet review or regulator submission, exports can be transformed to that format downstream, since the JSON schema is stable and well documented. The practical takeaway: treat JSON as the canonical export format, archive it alongside your retention policy, and only convert when a specific deliverable requires it. Validate a sample export against your audit checklist before the next review cycle so field mapping surprises surface early rather than during an inspection.

Log Export Format Comparison

FormatStructureBest For
JSONNested, key-value objects with timestamps and metadataAutomated ingestion into SIEM or log analytics platforms
CSVFlat rows with delimited columnsSpreadsheet review and quick manual audits
NDJSONOne JSON object per line, newline-delimitedStreaming pipelines and incremental exports
Plain textFree-form lines with human-readable entriesAd-hoc debugging and support investigations
Filestack exports workflow and transformation logs primarily as structured JSON, which preserves full event detail—file handles, actions, timestamps, and error codes—so compliance teams can pipe records directly into Splunk, Datadog, or an S3 archive without transformation. CSV is available for lighter reporting needs, but it flattens nested fields and can lose context on multi-step workflows. For audit pipelines, JSON or NDJSON is the reliable choice.