Microsoft Presidio is available as an add-on. If your workspace doesn’t have it enabled, choosing this detection method shows a Schedule consultation option instead of the configuration fields. The built-in regex method is always available. For the framework itself, see the Microsoft Presidio documentation.
How Presidio detection works
Presidio cleanly separates detection from anonymization, and MCP Manager runs both stages for you on each tool message:- Analyzer — identifies candidate PII and assigns each finding a confidence score, combining named-entity-recognition (NER) models, regular-expression recognizers, checksums, and surrounding-context cues.
- Anonymizer — takes the analyzer’s findings and transforms the matched spans according to the rule’s action.
US_DRIVER_LICENSE, DATE_TIME, MEDICAL_LICENSE, PERSON, LOCATION and NRP. Presidio ignores a match of one of these types inside a UUID (such as 7394320e in 7394320e-2f4d-11e6-9b3c-0242ac120002), for every rule and every action, even when the rule selects that type, so an ID is not masked into an invalid value. A real date or license number written in the shape of a UUID can therefore pass. Every other type, such as a credit card, SSN, phone number or bank account number, is detected inside a UUID as usual. The call’s log entry shows how many matches of each type were ignored. To redact UUIDs, add a regex rule.
Text that holds JSON
When a server returns JSON inside a text value, the analyzer reads the JSON’s strings decoded. A\n in the JSON is then a real line break, so a name on the line after it is found as it would be in plain text. Keys are screened too. A Replace rule changes only the values it tags and leaves the rest of the JSON as the server sent it.
- Dictionary words in field names. A key that looks like a field name can read as a date (
today), a place (US,Germany) or a nationality (american). ADATE_TIME,LOCATIONorNRPfound wholly inside such a key is ignored, under every action. Any other entity found in a key, such as an IBAN or a person’s name, is tagged or blocked as usual. - Numbers in the JSON. A
DATE_TIMEfound wholly inside a number of the JSON, such as an issue number1999or a timestamp, is ignored. Other entities in a number, such as a US SSN or a bank account number, are still found, with the key as context. - Detections across line breaks. The analyzer can run a name over a line break into the next line. A Replace rule tags each line of the detection on its own, so the line breaks stay. For a person’s name, a line with no letter or digit, such as a list’s
-, is left as it is. See Text that holds JSON for the details.
Entity types
In the rule editor you choose which entity types to detect. If you select none, all supported entity types are detected. The entity types MCP Manager exposes are Presidio’s standard global recognizers:- Financial —
CREDIT_CARD,IBAN_CODE,US_BANK_NUMBER,CRYPTO(wallet addresses) - Government and national IDs —
US_SSN,US_ITIN,US_PASSPORT,US_DRIVER_LICENSE,UK_NHS,MEDICAL_LICENSE - Contact and identity —
PERSON,EMAIL_ADDRESS,PHONE_NUMBER,LOCATION,NRP(nationality, religious, or political group) - Technical —
IP_ADDRESS,URL,DATE_TIME
Confidence threshold
The Confidence threshold is the minimum score — from 0.0 to 1.0 — an entity must reach to count as a detection. The default is 0.2. Lower values catch more entities but produce more false positives; higher values are more conservative. Because Presidio’s accuracy is tuning-dependent, treat the threshold as a dial: start at the default, watch what fires in your logs and alerts, and raise it if benign text is being flagged.Reliability with names and free-form text
Presidio’s reliability is not uniform across entity types, and the difference decides what you can safely trust it for. Structured identifiers —CREDIT_CARD, US_SSN, EMAIL_ADDRESS, IBAN_CODE, IP_ADDRESS, and the like — are found with patterns, checksums, and validation, so Presidio catches them dependably wherever they appear. Names and other context-dependent entities — PERSON, LOCATION, and NRP — are found by NLP models, which are markedly less reliable on short, terse, or loosely structured text. A person’s name in a task title, a column header, or a single-word field can be missed even when the same name in a full sentence would be caught.
Failure mode
Presidio runs as a service the gateway calls, so a Presidio rule has a Failure mode for when it’s unreachable, too slow, or errors:- Allow — let the message through unchanged. This is the default.
- Block — block the message as a precaution.
Actions
Presidio rules support two actions:- Block — block the whole message when any selected entity is detected; the anonymizer is not invoked.
- Replace — run the anonymizer to replace each detected entity with a tag naming its type. For example,
Customer email: alice@example.combecomesCustomer email: <EMAIL_ADDRESS>, and a card number becomes<CREDIT_CARD>. The surrounding text is left intact.
Presidio’s Replace tags each entity by type (
<EMAIL_ADDRESS>, <CREDIT_CARD>, …), which is more informative than the single <SENSITIVE> placeholder a regex Replace rule uses. The other regex actions — redact, mask, and hash — are not available for Presidio; use a regex rule if you need those.Example: broad PII protection
Detection method: Microsoft Presidio · Entity types: none (detect all) · Confidence threshold: 0.5 · Failure mode: Block · Action: Replace This catch-all rule tags every detected entity by type and fails closed if the service is down. Combine it with more targeted regex rules placed above it — for example, blocking prompt injection first, then replacing PII as a safety net. See Rule order.Good to know
Microsoft is candid about Presidio’s limits, and they apply here too:- No detection is guaranteed. Presidio uses automated detection, so it won’t catch every piece of sensitive data. Treat it as one layer, not a complete guarantee, and pair it with other controls for high-stakes data.
- Accuracy is tuning-dependent. Out of the box it’s strong on common PII, but the confidence threshold and entity selection materially change recall and the false-positive rate. Tune against your own traffic.
- Email detection checks the domain.
EMAIL_ADDRESSonly counts an address whose domain ends in a real top-level domain. An address such asname@acme.exampleorname@acme.testscores 0 and passes through untouched, whilename@acme.comis caught. Use real-looking domains in demo and test records, or the rule will look broken when it is not. - It’s a building block, not a full DLP product. Presidio complements enterprise data-loss-prevention tooling rather than replacing it.
- No per-call charges. Once the Presidio add-on is enabled on your plan, MCP Manager does not meter or charge per scan or per detection. Run Presidio rules on as much traffic as you need — there is no per-call or per-message fee for using it.
Further reading
Amazon Bedrock Guardrails
A managed guardrail for free-form text, added as a custom rule engine.
Lakera Guard
Classifier-based detection for names and free-form sensitive content.
Gateway Rules Overview
Detection methods, hooks, failure modes, actions, and rule ordering.
PII Filtering
Where Presidio fits in keeping customer PII out of the model.

