HAR Redactor
Create a reviewed HAR 1.2 export with explicit header, cookie, query and payload removal; strip unknown extensions while retaining useful capture measurements.
This tool processes its input contents in your browser instead of sending them to a processing service. Page, asset and fixed usage-measurement requests are described on the Privacy page.
Export a HAR 1.2 document using a visible removal policy for headers, cookies, query values and payloads, with explicit remaining-data warnings.
How to use it
- 1Paste or open a HAR 1.2 file and review the default removal policy.
- 2Redact locally and inspect the complete output plus removal counts.
- 3Download the HAR, then review remaining identifying data before sharing.
Example
- Input
- GET https://example.test/items?token=owned-example Authorization: Bearer owned-example
- Output
- GET https://example.test/items?token=%5BREDACTED%5D Authorization header removed
The path and hostname remain. The output needs human review and is not certified anonymous.
What happens to your data
HAR processing stays in the browser and never replays traffic. Removed values do not enter the count report. The export still contains deliberately retained capture data, so sharing remains a separate decision.
Last updated November 2026
Prepare a network capture for a deliberate review by removing selected sensitive fields locally. Chrome's sanitized HAR already omits common authorization/cookie headers. This tool adds query-value and payload removal, consistent request/query exports, and a report of what changed. The HAR Viewer remains available for a separate summary-only inspection.
How it works
Defaults remove Authorization, Proxy-Authorization, Cookie and Set-Cookie headers case-insensitively, remove request/response cookie arrays, drop request postData and response content text/encoding, and replace every URL query value with [REDACTED]. Request queryString is rebuilt from the resulting URL, preserving duplicate query names and their order. URL credentials and fragments are removed from request and redirect URLs. Redirects are not followed.
Advanced settings let you deliberately retain bodies/cookies or redact only named, case-sensitive query keys. Header names use literal HTTP token matching, with no regex. URLs can be normalized by the browser URL parser. Comments and unknown fields/extensions are dropped from known HAR 1.2 object structures, rather than copied blindly. The original timing, bodySize and headerSize observations remain measurements of the capture; they do not claim to describe the redacted bytes. At most 1,000 entries are processed.
Redaction is not anonymization or a safe-sharing guarantee. Hostnames, paths, IP addresses, page titles and remaining headers or standard metadata can still identify people, systems or credentials. Retained payloads can contain any data. Review the entire exported HAR with the intended recipient and policy in mind. No PII detector guesses which values are personal. The counts report never includes removed source values, though your chosen field names/settings remain visible. The tool does not replay traffic or contact captured hosts.
JSON input rejects duplicate keys and unpaired Unicode, with depth 32 and 20,000 values. Source input is limited to 500 KiB; JSON Schema and CSV schema inputs to 128 KiB. Complete output is limited to 512 KiB and has a shortened on-page preview. Copy and download contain the complete result. Editing source or settings immediately invalidates previous exports. Nothing is fetched or executed from the supplied text.
Worked example: HAR Redactor example
- Input
- GET https://example.test/items?token=owned-example Authorization: Bearer owned-example
- Result
- GET https://example.test/items?token=%5BREDACTED%5D Authorization header removed
The path and hostname remain. The output needs human review and is not certified anonymous.
Common use cases
- Reduce accidental credential exposure in a reviewed support capture.
- Strip bulky payload bodies while retaining capture measurements.
- Prepare a local debugging artifact without replaying its HTTP requests.
Frequently asked questions
Is the exported HAR safe to publish?+
No automatic guarantee is possible. Review hostnames, paths, metadata, remaining headers and any retained payloads before sharing.
Why are extension fields dropped?+
Unknown browser/vendor fields can carry sensitive values outside the selected policy. The initial tool omits and counts them rather than claiming they were inspected.
Does it call captured URLs?+
No. URLs are parsed as data only. No request is replayed, redirect followed or remote file fetched.
Standards and limitations
Related tools
SQL INSERT to CSV
Extract rows from a supported explicit-column INSERT VALUES subset, retain number spelling and distinguish SQL NULL from empty text.
JSON to TOML
Generate TOML from a strict JSON object, retain signed 64-bit integer digits, quote keys and reject unsupported null values.
JSON Merge Patch
Apply an object-shaped JSON merge patch, preview null deletions and whole-array replacement, and download the resulting JSON locally.
JSON Array Transformer
Sort JSON arrays by a property or remove duplicates while keeping exact numbers and the first matching item.