Communications compliance review

Step 01

Define and Align

Review representative examples and annotate the fields that need to change. Explain the edge cases to teach your team’s policy.

Output schema

DecisionChoose one: Whether the message can proceed under the supplied communications policy.
allow
escalate
IssuesChoose any: Policy issues found in the supplied message; empty when none apply.
guaranteed_outcome
missing_qualification
sensitive_data
ExplanationText: Specific message evidence supporting the decision.

app.sutro.sh/functions/communications-compliance-review

Collecting feedback · 1/3

Review the external message against the supplied communications policy and source facts.

Example 1 of 3

Estimate rewritten as a guarantee

Model input

Policy: Synthetic communications policy. External messages must not promise guaranteed outcomes. Preliminary estimates must be labeled as estimates and include the stated dependency. Flag customer identifiers as sensitive data. Escalate messages with either a prohibited claim or sensitive data; otherwise allow.
Source facts: Preliminary delivery estimate: Friday, dependent on completion of validation.
Message: We guarantee delivery by Friday.

Model output

Decision

IssuesMultiple values

Explanation

Why? Optional guidance

Step 02

Model Sweep

Compare models on the same expert annotations, then choose a quality–cost tradeoff.

Compare models on expert agreement and cost. The interactive chart uses illustrative measurements.

Illustrative sweep · select Value or Best to use that model in your function.

Step 03

Deployment

Call Sutro Functions like any other software. Swap the model as soon as something more performant is released.

Selected modelWaiting for sweep results.

Real-time

import sutro as so

input_data = {
    "policy": "Synthetic communications policy. External messages must not promise guaranteed outcomes. Preliminary estimates must be labeled as estimates and include the stated dependency. Flag customer identifiers as sensitive data. Escalate messages with either a prohibited claim or sensitive data; otherwise allow.",
    "source_facts": "Preliminary delivery estimate: Friday, dependent on completion of validation.",
    "message": "We guarantee delivery by Friday."
}

result = so.run_function(
    "communications-compliance-review",
    input_data,
)

curl --fail-with-body -X POST "${SUTRO_API_URL%/}/v1/functions/communications-compliance-review/run" \
  -H "Authorization: Key $SUTRO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "input": {
    "policy": "Synthetic communications policy. External messages must not promise guaranteed outcomes. Preliminary estimates must be labeled as estimates and include the stated dependency. Flag customer identifiers as sensitive data. Escalate messages with either a prohibited claim or sensitive data; otherwise allow.",
    "source_facts": "Preliminary delivery estimate: Friday, dependent on completion of validation.",
    "message": "We guarantee delivery by Friday."
  }
}'

Batch

import sutro as so

job_id = so.batch_run_function(
    name="communications-compliance-review",
    data="communications_compliance_review.csv",
)
results = so.await_job_completion(job_id)

curl --fail-with-body -X POST "${SUTRO_API_URL%/}/v1/batch-inference" \
  -H "Authorization: Key $SUTRO_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "communications-compliance-review",
  "inputs": [
    {
      "policy": "Synthetic communications policy. External messages must not promise guaranteed outcomes. Preliminary estimates must be labeled as estimates and include the stated dependency. Flag customer identifiers as sensitive data. Escalate messages with either a prohibited claim or sensitive data; otherwise allow.",
      "source_facts": "Preliminary delivery estimate: Friday, dependent on completion of validation.",
      "message": "We guarantee delivery by Friday."
    }
  ]
}'