DeepRead Form Fill

PassAudited by ClawScan on May 1, 2026.

Overview

The skill does what it advertises—sending user-selected PDFs and JSON data to DeepRead to fill forms—but users should be aware it involves sensitive documents, an API key, and external hosted results.

Install only if you are comfortable sending the relevant PDFs and JSON form data to DeepRead's external API. Use a protected API key, verify the package/version, review the service's privacy and retention terms, and check completed forms before using or submitting them.

Findings (3)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

Anyone with the API key could potentially use the user's DeepRead account or quota.

Why it was flagged

The skill requires a DeepRead API key and sends it as an API authentication header. This is expected for the service, but the key grants access to the user's DeepRead account/API quota.

Skill content
export DEEPREAD_API_KEY="sk_live_your_key_here" ... Authentication: `X-API-Key` header (required)
Recommendation

Use a dedicated, rotatable API key if possible, store it securely, and revoke it if it is exposed.

What this means

Private forms and filled data may be processed by a third-party service and made available through a signed download link.

Why it was flagged

The documented workflow uploads PDFs and potentially sensitive form fields to DeepRead, then returns a hosted download URL. This is core to the skill, but it means sensitive data leaves the local environment.

Skill content
curl -X POST https://api.deepread.tech/v1/form-fill ... -F "file=@tax_form.pdf" ... "ssn": "123-45-6789" ... "filled_form_url": "https://storage.deepread.tech/form_fill/.../filled.pdf"
Recommendation

Review DeepRead's privacy and retention terms, avoid uploading unnecessary sensitive data, use shorter URL expirations where available, and only use trusted webhook endpoints.

What this means

A user may have less assurance that the reviewed files exactly match the intended published release.

Why it was flagged

The registry metadata lists version 1.1.0 and an unknown source, while included package metadata files show version 1.0.0. This is a packaging/provenance consistency issue, not evidence of malicious behavior.

Skill content
Version: 1.1.0 ... Source: unknown
Recommendation

Verify the publisher, homepage/repository, and version before relying on the skill for sensitive documents.