DataFog MCP · Now available ↗

Check sensitive files
before your agent opens them.

Find personal data and credentials locally. Create redacted copies without returning matched values to the agent.

Get DataFog MCP ↗ Build with DataFog Core ↓

Get started with Claude Code

uv tool install datafog-mcp

Choose your allowed folders, then connect Claude Code. Follow the setup guide ↗

DataFog MCP + Claude Code
The engine behind DataFog MCP

Local PII protection
for your application.

Build directly with DataFog Core. Detect sensitive information in text and structured records, then choose how to protect it before passing it to a model, a log, or another system.

SDKs for Rust, Python, Node.js, and browsers.

DataFog Core · Latest release ↗ 
python -m pip install datafog-core

No separate service to deploy. Installation guide ↗

Open sourceMIT licensed Built in RustOne shared privacy engine
Local detectionRuns inside your application
Ready out of the boxStart detecting with the built-in engine
Detection is the beginning

Choose what happens
to sensitive data.

Different workflows need different protections. Select the information to transform, apply rules by entity type, and exempt approved values.

Redact · Mask · Remove

Remove what you don’t need.

Replace sensitive values with typed placeholders, mask them with optional partial visibility, or remove them from the text.

Explore transformations ↗

Pseudonymize

Preserve useful relationships.

Use keyed pseudonymization when repeated values need consistent replacements. Your application supplies the key provider.

Understand pseudonymization ↗

Tokenize · Restore

Restore when authorized.

Use provider-backed tokenization for reversible replacements. Your provider owns storage and authorization; Core checks the request scope.

See tokenization and restoration ↗

Redaction, masking, and removal work across all SDKs. Pseudonymization and tokenization use application-supplied providers in Rust, Python, and Node.js.

Build with DataFog

Start in the runtime
you already use.

Scan for findings, transform selected values, or combine both in one operation. The same Core defines the privacy behavior across SDKs.

Installation guide ↗

Already using the datafog Python package? Core has a separate package and API. Follow the migration guide.

# pip install datafog-core
from datafog_core import scan_and_transform

result = scan_and_transform(
    "Contact ada@example.invalid",
    {"transform": {"default": {"strategy": "redact"}}},
)
print(result.text)
# Contact [EMAIL]
Python reference ↗
Use cases

Protect data wherever it goes next.

Check sensitive information before it enters an agent workflow, becomes a shared dataset, or leaves your environment.

Agent workflows

Keep PII out of agent context.

Scan and redact inputs before they reach your agent. Apply the same checks to tool calls and trace payloads before they are sent or stored.

Data preparation

Prepare data for its next use.

Create protected copies of text and structured records for analysis, testing, or downstream processing. Remove sensitive values before sharing data with teammates or third parties.

Outbound protection

Check before PII leaves.

Scan outgoing payloads before an API request, upload, or export. Transform detected PII or use the findings to stop the transfer in your application.

Where we’re going

One privacy engine.
More places to use it.

We’re working toward consistent detection and transformation across SDKs, agents, and gateways. Core provides the shared engine. Integrations connect it to workflows and decide when to allow, warn, block, or transform.

Available in Core today

Build the protection you need.

  • Text scanning and structured-record protection, including supported person-name fields.
  • SDKs for Rust, Python, Node.js, and browsers.
  • Configurable transformations, with provider-backed operations in server runtimes.

Next direction

Bring the engine to more workflows.

  • Adopt Core in downstream integrations.
  • Carry shared privacy behavior into more agent and gateway workflows.
  • Keep workflow-specific interception and enforcement in the integrations.

Read the Core 0.3.0 release notes ↗ · Follow the Core roadmap ↗

Performance varies by workload. See the 0.3.0 benchmark methodology and results.

Build privacy protection
into your next application.

Start with a scan, choose a transformation, and connect DataFog to your workflow.

Get started ↗Read the documentation ↗

Contact us

Have a workflow in mind? Talk to us about protecting sensitive data in your application.

Email hi@datafog.ai ↗