PII Detection

kiteguard scans for personally identifiable information (PII) across three interception points:

UserPromptSubmit — the user's own prompt
PostToolUse — content returned from files or web pages Claude reads
Stop — the final assistant response before delivery

Configuration

pii:
  enabled: true
  block_on_prompt: false   # redact from prompt, don't block
  block_on_response: true  # block if PII found in Claude's response
  types:
    - ssn
    - credit_card
    - email
    - phone_us
    - passport

Fields

Field	Default	Description
`enabled`	`true`	Master toggle for all PII detection
`block_on_prompt`	`false`	Block (exit 2) when PII found in user prompt
`block_on_response`	`true`	Block response delivery when PII found in final response
`types`	all	List of PII types to detect (omit a type to disable it)

Supported PII types

Type	Example	Pattern description
`ssn`	`123-45-6789`	US Social Security Number (dashes or dots)
`credit_card`	`4111 1111 1111 1111`	Visa, Mastercard, Amex, Discover
`email`	`alice@example.com`	Standard email address
`phone_us`	`(555) 867-5309`	US phone number (multiple formats)
`passport`	`A12345678`	US passport number

block_on_prompt vs block_on_response

Setting	Effect
`block_on_prompt: false`	PII in the user's prompt is logged but not blocked. Claude processes it — useful when the user intentionally pastes data to process.
`block_on_prompt: true`	Blocks the request entirely. The user must remove PII before Claude sees it.
`block_on_response: true`	Blocks Claude's final reply if it contains PII — prevents Claude from echoing sensitive data back.

Notes

All regex patterns are anchored to word boundaries to reduce false positives.
Credit card detection uses a simplified Luhn-adjacent regex — it catches common formats but is not an authoritative validator.
Prompt text is never stored in the audit log, only its hash.