Redacting sensitive content

AI agents routinely handle sensitive information — payment details, social security numbers, account credentials, names and addresses. Content redaction masks that information in everything SignalWire records or transmits about the call: logs, events, webhook payloads, the call timeline, and the post-conversation call log.

The conversation itself is untouched. The caller hears the agent normally, and the agent understands the caller perfectly — only the records change.

Enable redaction

Turn redaction on with a single parameter, redact_prompt, in the ai method’s params block:

1
version: 1.0.0
2
sections:
3
  main:
4
    - answer: {}
5
    - ai:
6
        prompt:
7
          text: You are a payments support agent. Help callers update their billing details.
8
        params:
9
          redact_prompt: credit card numbers, CVVs, social security numbers, and full names

The value of redact_prompt does two jobs: it switches redaction on, and it describes — in plain language — what counts as sensitive. Anything matching that description is replaced with ---- wherever the call is recorded or transmitted.

How it works

Redaction covers both sides of the conversation. Your redact_prompt description guides the agent to treat matching content as sensitive whenever it speaks — the first time it says it, when it repeats it back, when it confirms it. What the caller says is masked separately, before it is stored or delivered. In both cases the real words still flow through the live conversation; only the records change.

With redaction on, here is what each surface shows:

Redaction is performed by AI, not by a fixed pattern-matcher. It errs on the side of masking too much rather than too little, but it can occasionally miss — treat it as a strong safeguard for your logs and integrations rather than an absolute guarantee.

Keep it fast

Redaction runs as a lightweight background task while the caller is waiting for a response. Two companion parameters keep that work quick.

utility_model selects the model used for background tasks like redaction and transcription cleanup. It defaults to the agent’s main model, which is usually larger and slower than these tasks need.

auto_correct cleans up the transcription of the caller’s speech — converting spoken numbers to digits, formatting addresses and phone numbers, and fixing obvious mishearings. When used alongside redact_prompt, cleanup and redaction happen together in a single step instead of two.

A complete configuration:

1
version: 1.0.0
2
sections:
3
  main:
4
    - answer: {}
5
    - ai:
6
        prompt:
7
          text: You are a payments support agent. Help callers update their billing details.
8
        params:
9
          redact_prompt: credit card numbers, CVVs, social security numbers, and full names
10
          utility_model: gpt-4o-mini
11
          auto_correct: true
12
          enable_text_normalization: "off"

Verify redaction

Limitations

• It is not model-input privacy. The AI model receives the real text every turn. Redaction keeps sensitive data out of your logs, webhooks, and call records — keeping it away from the model is a different requirement that redaction does not satisfy.
• It is best-effort. Redaction is AI-driven and biased toward over-masking, but a value can occasionally slip through. Sharpen the redact_prompt description if a category leaks.
• It adds a small amount of background work per turn. Point utility_model at a small, fast model, and enable auto_correct to fold cleanup and redaction into one step.
• It applies to text records, not audio. The caller hears everything in full, and if you record calls, the recording still contains the real spoken audio.

Next steps